metadata

language:
  - en
license: apache-2.0
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:6300
  - loss:MatryoshkaLoss
  - loss:MultipleNegativesRankingLoss
base_model: BAAI/bge-base-en-v1.5
widget:
  - source_sentence: What are the main components of technology and infrastructure costs?
    sentences:
      - >-
        As of January 29, 2023, from the total aggregate lease obligations of
        $14.7 billion, $1.5 billion was payable within 12 months.
      - >-
        Technology and infrastructure costs include payroll and related expenses
        for employees involved in the research and development of new and
        existing products and services, development, design, and maintenance of
        our stores, curation and display of products and services made available
        in our online stores, and infrastructure costs.
      - >-
        'Note 13 — Commitments and Contingencies — Litigation and Other Legal
        Matters' is stated to be part of Part IV, Item 15 of the consolidated
        financial statements within an Annual Report on Form 10-K.
  - source_sentence: >-
      How is Meta's workforce comprised in terms of diversity as of December 31,
      2022?
    sentences:
      - >-
        As of December 31, 2022, our global employee base was composed of 45.4%
        underrepresented people, with 47.9% underrepresented people in the U.S.,
        and 43.1% of our leaders in the U.S. being people of color.
      - >-
        IBM's 2023 Annual Report to Stockholders includes the Financial
        Statements and Supplementary Data on pages 44 through 121.
      - >-
        Factors affecting the overall effective tax rate include acquisitions,
        changes in corporate structures, location of business functions, the mix
        and amount of income, agreements with tax authorities, and variations in
        estimated and actual pre-tax income.
  - source_sentence: >-
      What was the valuation allowance against deferred tax assets at the end of
      2023, and what changes may affect its realization?
    sentences:
      - >-
        At December 31, 2020, valuation allowances against deducted assets were
        $7.0 billion. The ability to realize deductible benefits in future is
        contingent on producing any estimated sufficient values in
        cash-generating, with effects are modifications in trade situations,
        political of force, or those actions meaningfully impacting on the
        values.
      - >-
        Amazon considers its intellectual property essential for its success,
        utilizing trademark, copyright, and patent law, trade-secret protection,
        and confidentiality and/or license agreements to protect these rights.
      - >-
        During 2023, AMC served as the theatrical distributor for two theatrical
        releases: TAYLOR SWIFT | THE ERAS TOUR and RENAISSANCE: A FILM BY
        BEYONCÉ.
  - source_sentence: >-
      What significant services are included in Iron Mountain's service
      revenues?
    sentences:
      - >-
        The decrease in net income in 2022 was primarily due to an increase in
        selling, general and administrative expenses of $532.4 million, an
        impairment charge recognized in 2022 of $407.9 million, an increase in
        income tax expense of $119.2 million, partially offset by an increase in
        gross profit of $883.8 million, a decrease in acquisition-related
        expenses of $41.4 million, a gain on disposal of assets of $10.2
        million, and an increase in other income (expense), net of $3.6 million.
      - >-
        Service revenues include charges for the handling of records,
        destruction services, digital solutions, and data center services.
      - >-
        The total operating expenses for Chipotle Mexican Grill in 2023 amounted
        to $8,313,836.
  - source_sentence: >-
      In which part and item of the Annual Report on Form 10-K can the
      consolidated financial statements be found?
    sentences:
      - >-
        In order to maintain leadership, we optimize our portfolio with organic
        and inorganic innovations and effective resource allocation. These
        investments not only drive current performance but will extend our
        innovation leadership into the future.
      - >-
        Our Consumer Wireline business unit offers AT&T Internet Air, which is a
        fixed wireless access product that provides home internet services
        delivered over our 5G wireless network where available.
      - >-
        The consolidated financial statements and accompanying notes listed in
        Part IV, Item 15(a)(1) of this Annual Report on Form 10-K are included
        elsewhere in this Annual Report on For... 10-K.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@3
  - cosine_accuracy@5
  - cosine_accuracy@10
  - cosine_precision@1
  - cosine_precision@3
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@1
  - cosine_recall@3
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@10
  - cosine_mrr@10
  - cosine_map@100
model-index:
  - name: BGE base Financial Matryoshka
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 768
          type: dim_768
        metrics:
          - type: cosine_accuracy@1
            value: 0.7114285714285714
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.8371428571428572
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.87
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.9057142857142857
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.7114285714285714
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.27904761904761904
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.174
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.09057142857142855
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.7114285714285714
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.8371428571428572
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.87
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.9057142857142857
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.8110932340412786
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.7804977324263039
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.784240984630403
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 512
          type: dim_512
        metrics:
          - type: cosine_accuracy@1
            value: 0.7157142857142857
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.83
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.87
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.9071428571428571
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.7157142857142857
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.27666666666666667
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.174
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.0907142857142857
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.7157142857142857
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.83
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.87
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.9071428571428571
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.8116485651477514
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.7810300453514737
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.7845397715740386
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@1
            value: 0.7128571428571429
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.8214285714285714
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.86
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.9042857142857142
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.7128571428571429
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.27380952380952384
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.17199999999999996
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.09042857142857143
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.7128571428571429
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.8214285714285714
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.86
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.9042857142857142
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.8071701520591847
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.7762494331065761
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.7797123012827435
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 128
          type: dim_128
        metrics:
          - type: cosine_accuracy@1
            value: 0.71
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.81
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.8442857142857143
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.8985714285714286
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.71
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.27
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.16885714285714284
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.08985714285714284
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.71
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.81
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.8442857142857143
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.8985714285714286
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.801264041144764
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.7705725623582764
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.7744092505881914
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 64
          type: dim_64
        metrics:
          - type: cosine_accuracy@1
            value: 0.6685714285714286
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.78
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.8257142857142857
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.8757142857142857
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.6685714285714286
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.25999999999999995
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.16514285714285715
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.08757142857142856
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.6685714285714286
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.78
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.8257142857142857
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.8757142857142857
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.7698003192070297
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.7363242630385484
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.7409337390692949
            name: Cosine Map@100

BGE base Financial Matryoshka

This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5 on the json dataset. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: BAAI/bge-base-en-v1.5
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- json
Language: en
License: apache-2.0

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("TatvaRA/bge-base-financial-matryoshka")
# Run inference
sentences = [
    'In which part and item of the Annual Report on Form 10-K can the consolidated financial statements be found?',
    'The consolidated financial statements and accompanying notes listed in Part IV, Item 15(a)(1) of this Annual Report on Form 10-K are included elsewhere in this Annual Report on For... 10-K.',
    'Our Consumer Wireline business unit offers AT&T Internet Air, which is a fixed wireless access product that provides home internet services delivered over our 5G wireless network where available.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Information Retrieval

Dataset: dim_768
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 768
}
```

Metric	Value
cosine_accuracy@1	0.7114
cosine_accuracy@3	0.8371
cosine_accuracy@5	0.87
cosine_accuracy@10	0.9057
cosine_precision@1	0.7114
cosine_precision@3	0.279
cosine_precision@5	0.174
cosine_precision@10	0.0906
cosine_recall@1	0.7114
cosine_recall@3	0.8371
cosine_recall@5	0.87
cosine_recall@10	0.9057
cosine_ndcg@10	0.8111
cosine_mrr@10	0.7805
cosine_map@100	0.7842

Information Retrieval

Dataset: dim_512
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 512
}
```

Metric	Value
cosine_accuracy@1	0.7157
cosine_accuracy@3	0.83
cosine_accuracy@5	0.87
cosine_accuracy@10	0.9071
cosine_precision@1	0.7157
cosine_precision@3	0.2767
cosine_precision@5	0.174
cosine_precision@10	0.0907
cosine_recall@1	0.7157
cosine_recall@3	0.83
cosine_recall@5	0.87
cosine_recall@10	0.9071
cosine_ndcg@10	0.8116
cosine_mrr@10	0.781
cosine_map@100	0.7845

Information Retrieval

Dataset: dim_256
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 256
}
```

Metric	Value
cosine_accuracy@1	0.7129
cosine_accuracy@3	0.8214
cosine_accuracy@5	0.86
cosine_accuracy@10	0.9043
cosine_precision@1	0.7129
cosine_precision@3	0.2738
cosine_precision@5	0.172
cosine_precision@10	0.0904
cosine_recall@1	0.7129
cosine_recall@3	0.8214
cosine_recall@5	0.86
cosine_recall@10	0.9043
cosine_ndcg@10	0.8072
cosine_mrr@10	0.7762
cosine_map@100	0.7797

Information Retrieval

Dataset: dim_128
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 128
}
```

Metric	Value
cosine_accuracy@1	0.71
cosine_accuracy@3	0.81
cosine_accuracy@5	0.8443
cosine_accuracy@10	0.8986
cosine_precision@1	0.71
cosine_precision@3	0.27
cosine_precision@5	0.1689
cosine_precision@10	0.0899
cosine_recall@1	0.71
cosine_recall@3	0.81
cosine_recall@5	0.8443
cosine_recall@10	0.8986
cosine_ndcg@10	0.8013
cosine_mrr@10	0.7706
cosine_map@100	0.7744

Information Retrieval

Dataset: dim_64
Evaluated with InformationRetrievalEvaluator with these parameters:
```
{
    "truncate_dim": 64
}
```

Metric	Value
cosine_accuracy@1	0.6686
cosine_accuracy@3	0.78
cosine_accuracy@5	0.8257
cosine_accuracy@10	0.8757
cosine_precision@1	0.6686
cosine_precision@3	0.26
cosine_precision@5	0.1651
cosine_precision@10	0.0876
cosine_recall@1	0.6686
cosine_recall@3	0.78
cosine_recall@5	0.8257
cosine_recall@10	0.8757
cosine_ndcg@10	0.7698
cosine_mrr@10	0.7363
cosine_map@100	0.7409

Training Details

Training Dataset

json

Dataset: json
Size: 6,300 training samples
Columns: anchor and positive
Approximate statistics based on the first 1000 samples:
anchor positive
type string string
details
min: 9 tokens
mean: 20.16 tokens
max: 51 tokens

min: 4 tokens
mean: 45.99 tokens
max: 281 tokens

	anchor	positive
type	string	string
details	min: 9 tokens mean: 20.16 tokens max: 51 tokens	min: 4 tokens mean: 45.99 tokens max: 281 tokens

Samples:

anchor	positive
`What percentage of total revenues did STELARA account for in fiscal 2023 for the Company?`	`Sales of the Company’s largest product, STELARA (ustekinumab), accounted for approximately 12.8% of the Company's total revenues for fiscal 2023.`
`What is the effective date for the new accounting standard ASU No. 2022-04 regarding liabilities in supplier finance programs?`	`In September 2022, the FASB issued ASU No. 2022-04, “Liabilities—Supplier Finance Programs (Topic 405-50) - Disclosure of Supplier Finance Program Obligations,” which is effective for fiscal years beginning after December 15, 2022, including interim periods within those fiscal years.`
`What was the pre-tax net favorable prior period development for 2022 and what factors contributed to it?`	`The pre-tax net favorable prior period development for 2022 was $876 million. Adverse development factors like molestation claims, primarily reviver statute-related compromising $155 million, and $113 million related to legacy asbestos and environmental exposures significantly influenced this outcome.`

Loss: MatryoshkaLoss with these parameters:

{
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        768,
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: epoch
per_device_train_batch_size: 32
per_device_eval_batch_size: 16
gradient_accumulation_steps: 16
learning_rate: 2e-05
num_train_epochs: 4
lr_scheduler_type: cosine
warmup_ratio: 0.1
fp16: True
load_best_model_at_end: True
optim: adamw_torch_fused
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: epoch
prediction_loss_only: True
per_device_train_batch_size: 32
per_device_eval_batch_size: 16
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 16
eval_accumulation_steps: None
learning_rate: 2e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 4
max_steps: -1
lr_scheduler_type: cosine
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
use_ipex: False
bf16: False
fp16: True
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: True
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch_fused
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: False
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
dispatch_batches: None
split_batches: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	dim_768_cosine_ndcg@10	dim_512_cosine_ndcg@10	dim_256_cosine_ndcg@10	dim_128_cosine_ndcg@10	dim_64_cosine_ndcg@10
0.8122	10	1.6789	-	-	-	-	-
0.9746	12	-	0.7976	0.8019	0.7944	0.7781	0.7387
1.6244	20	0.6377	-	-	-	-	-
1.9492	24	-	0.8071	0.8080	0.8016	0.7940	0.7594
2.4365	30	0.5295	-	-	-	-	-
2.9239	36	-	0.8110	0.8122	0.8067	0.8000	0.7697
3.2487	40	0.4367	-	-	-	-	-
3.8985	48	-	0.8111	0.8116	0.8072	0.8013	0.7698

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.11.12
Sentence Transformers: 4.1.0
Transformers: 4.41.2
PyTorch: 2.1.2+cu121
Accelerate: 1.5.2
Datasets: 2.19.1
Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning},
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}