Dataset Viewer
conversations
stringlengths 2.72k
3.29k
|
|---|
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me funds with high dividend yield'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/yield/dividend", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me dividend yield mutual funds'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/yield/dividend", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/dividend_yield_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high alpha and low volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/alpha", "qualifier": "high", "time": null, "quantifier": null}, {"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me funds with high returns and low volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns", "qualifier": "high", "time": null, "quantifier": null}, {"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'recommend equity funds with low volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [{"node": "asset_type/equity"}], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me funds with high alpha and low volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/alpha", "qualifier": "high", "time": null, "quantifier": null}, {"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high risk-adjusted returns'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/risk_adjusted_returns", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'small cap funds with high volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/style/size/small_cap", "qualifier": null, "time": null, "quantifier": null}, {"node": "attribute/technical/risk/volatility", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/small_cap_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me mutual funds focused on telecom services'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [], "exposures": [{"node": "exposure/sector/communication_services/telecommunication_services", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/sectoral_/_thematic_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'low volatility funds'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with low volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high annualized returns from last year'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/annualized_returns", "qualifier": "high", "time": "1y", "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me funds with high alpha returns focusing on IT sector'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/alpha", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/sector/information_technology", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'list funds with high dividend yields'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/yield/dividend", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'looking for large cap funds with annualized returns over 15%'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/style/size/large_cap", "qualifier": "high", "time": null, "quantifier": null}, {"node": "attribute/technical/returns/annualized_returns", "qualifier": null, "time": null, "quantifier": ">15%"}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/large_cap_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'recommend funds with low volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high dividend yield and low volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/yield/dividend", "qualifier": "high", "time": null, "quantifier": null}, {"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me funds with high alpha'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/alpha", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me funds with low beta over the last year'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/beta", "qualifier": "low", "time": "1y", "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'best small cap funds with low volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/performance", "qualifier":"high", "time": null, "quantifier": null},{"node": "attribute/style/size/small_cap", "qualifier":"high", "time": null, "quantifier": null}, {"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/small_cap_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'worst small cap funds with low volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/performance", "qualifier":"low", "time": null, "quantifier": null},{"node": "attribute/style/size/small_cap", "qualifier":"high", "time": null, "quantifier": null}, {"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/small_cap_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high momentum exposure'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [], "exposures": [{"node": "exposure/factor/momentum", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me funds in the tech sector'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [], "exposures": [{"node": "exposure/sector/information_technology", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/sectoral_/_thematic_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'find funds with high alpha'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/alpha", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'Show me funds with high dividend yield.'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/yield/dividend", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'low volatility funds with high momentum'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/risk/volatility", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/factor/momentum", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me funds with midcap exposure and low volatility'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/style/size/midcap", "qualifier":"high", "time": null, "quantifier": null},{"node": "attribute/technical/risk", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/factor/size", "qualifier": "high", "quantifier": null}, {"node": "exposure/factor/volatility", "qualifier": "low", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/mid_cap_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'find funds with high relative volume trend'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/volume/relative_volume", "qualifier": "high", "time": null, "quantifier": null}, {"node": "attribute/technical/trend", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'best alpha funds'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/alpha", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'looking for funds with high alpha and tech exposure'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/alpha", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/sector/information_technology", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with good credit rating'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/credit_rating", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high alpha'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/alpha", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high alpha and telecommunication exposure'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/alpha", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/sector/communication_services/telecommunication_services", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high beta and small cap focus'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/beta", "qualifier": "high", "time": null, "quantifier": null}, {"node": "attribute/style/size/small_cap", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/small_cap_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high dividend yield'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/yield/dividend", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high credit rating'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/credit_rating", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'find funds with strong momentum'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/trend/up_trend", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/factor/momentum", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'best dividend yield funds'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/yield/dividend", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/dividend_yield_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'high beta funds'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/beta", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high annualized returns'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/annualized_returns", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'pharma funds with high returns'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/sector/health_care/pharmaceuticals_biotechnology_and_life_sciences", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/sectoral_/_thematic_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'small cap funds'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/style/size/small_cap", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/small_cap_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'small cap etfs'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/style/size/small_cap", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/etf"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds in IT sector with high momentum'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [], "exposures": [{"node": "exposure/sector/information_technology", "qualifier": "high", "quantifier": null}, {"node": "exposure/factor/momentum", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'large cap funds with high return'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/style/size/large_cap", "qualifier": "high", "time": null, "quantifier": null}, {"node": "attribute/technical/returns", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/large_cap_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'debt funds with high credit rating'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/credit_rating", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/debt_schemes"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me high return tech funds'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/sector/information_technology", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/sectoral_/_thematic_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'real estate funds with high returns'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/sector/real_estate", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/sectoral_/_thematic_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me mid-cap funds with high credit rating'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/credit_rating", "qualifier": "high", "time": null, "quantifier": null}, {"node": "attribute/style/size/mid_cap", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/mid_cap_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'ETFs with high momentum'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [], "exposures": [{"node": "exposure/factor/momentum", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/etf"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'low-risk funds in consumer discretionary'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/risk", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/sector/consumer_discretionary", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/sectoral_/_thematic_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'suggest funds with a high credit rating and exposure to Nifty 50'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/credit_rating", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [{"node": "exposure/index/nifty_50", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/other/index_funds_etfs"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'high-yield mutual funds with low credit risk'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/yield/dividend", "qualifier": "high", "time": null, "quantifier": null}, {"node": "attribute/fundamental/credit_rating", "qualifier": "low", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'high credit rating funds'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/credit_rating", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'Show me mutual funds with over 25% exposure to FMCG'}, {'from': 'gpt', 'value': '{ "parsed_output": { "attributes": [], "exposures": [{ "node": "exposure/sector/consumer_staples", "qualifier": null, "quantifier": ">25%" }], "tickers": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/sectoral_/_thematic_fund"}], "vehicles": [{ "node": "vehicle/funds/mutual_fund" }] } }'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high vwap'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/price/vwap", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'suggest funds with high momentum exposure'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [], "exposures": [{"node": "exposure/factor/momentum", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'top dividend yield funds'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/fundamental/yield/dividend", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/dividend_yield_fund"}], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'funds with high momentum exposure and annualized returns over 10%'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/annualized_returns", "qualifier": "high", "time": null, "quantifier": ">10%"}], "exposures": [{"node": "exposure/factor/momentum", "qualifier": "high", "quantifier": null}], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'show me funds with high annualized returns'}, {'from': 'gpt', 'value': '{"parsed_output": {"attributes": [{"node": "attribute/technical/returns/annualized_returns", "qualifier": "high", "time": null, "quantifier": null}], "exposures": [], "tickers": [], "amc": [], "asset_types": [], "sebi": [], "vehicles": [{"node": "vehicle/funds/mutual_fund"}], "objectives": []}}'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'Show me FMCG funds'}, {'from': 'gpt', 'value': '{ "parsed_output": { "attributes": [], "exposures": [ { "node": "exposure/sector/consumer_staples", "qualifier": "high", "quantifier": null } ], "tickers": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/sectoral_/_thematic_fund"}], "vehicles": [] } }'}] |
[{'from': 'system', 'value': '\nYou are a semantic parser specializing in financial queries. Your task is analyze financial queries and extract relevant information into a structured format.\n\n# Output Schema\n\n{\n "attributes": [{"node": string, "qualifier": string|null, "time": string|null, "quantifier": string|null}],\n "exposures": [{"node": string|null, "qualifier": string|null, "quantifier": string|null}],\n "tickers": [{"name": string}],\n "asset_types": [{"node": string}],\n "sebi": [{"node": string}],\n "vehicles": [{"node": string}]\n}\n\n\n# Guidelines\n\n1. Attributes:\n - Use full hierarchical paths (e.g., \'attribute/technical/returns/annualized_returns\')\n - For queries related to large-cap, mid-cap, small-cap, use \'attribute/style/size/large_cap\', \'attribute/style/size/mid_cap\', \'attribute/style/size/small_cap\'\n - Qualifier: "high", "low", or null\n - Time: Use \'d\', \'w\', \'m\', \'y\', \'ytd\' units. For recent/current, use \'1w\'\n - Quantifier format: <arithmetic_operator><value><unit>\n\n2. Exposures:\n - Use full sector paths (e.g., \'exposure/sector/financials\')\n - Use \'exposure/ticker/security_name\' when there is a ticker which is being queried as holdings/compositions of the main security.\n - Qualifier: "high" (default), "low", or null\n - Quantifier format: <arithmetic_operator><value>% (e.g., ">5%")\n - Tickers - Exposure tickers refers to the list of tickers that are being queried as holdings/compositions of the main security.\n\n3. Tickers:\n - Include any mentioned securities names (e.g., \'Nippon India Nifty smallcap\', \'Tata\', \'Edelweiss\')\n - Use exact names as mentioned in the query\n\n4. Asset Types:\n - Use full paths (e.g., \'asset_type/equity\', \'asset_type/bonds\')\n - If stock or equity fund is mentioned, use \'asset_type/equity\'\n\n5. SEBI:\n - Use full classification paths (e.g., \'sebi/equity_schemes/large_cap_fund\')\n - For sector or theme-specific requests, use \'sebi/equity_schemes/sectoral_/_thematic_fund\'\n - For tax-saving funds, use \'sebi/equity_schemes/elss\'\n\n6. Vehicles:\n - Use full paths (e.g., \'vehicle/funds/mutual_fund\', \'vehicle/funds/etf\')\n - Always include if a specific investment vehicle is mentioned or implied\n\n# Important Notes:\n- Today\'s date: 2025-05-28\n- For qualitative performance: use \'attribute/technical/performance\'\n- Be precise with quantifiers\n- Provide only the structured output, no additional text\n'}, {'from': 'human', 'value': 'Help me invest in FMCG'}, {'from': 'gpt', 'value': '{ "parsed_output": { "attributes": [], "exposures": [ { "node": "exposure/sector/consumer_staples", "qualifier": "high", "quantifier": null } ], "tickers": [], "asset_types": [], "sebi": [{"node": "sebi_classification/equity_schemes/sectoral_/_thematic_fund"}], "vehicles": [] } }'}] |
End of preview. Expand
in Data Studio
README.md exists but content is empty.
- Downloads last month
- 78