Ksenia Se's picture

1 92 3

Ksenia Se

Kseniase

·

https://www.turingpost.com/

AI & ML interests

None yet

Recent Activity

reacted to their post with 👍 2 days ago

16 new research on inference-time scaling: For the last couple of weeks a large amount of studies on inference-time scaling has emerged. And it's so cool, because each new paper adds a trick to the toolbox, making LLMs more capable without needing to scale parameter count of the models. So here are 13 new methods + 3 comprehensive studies on test-time scaling: 1. https://huggingface.co/papers/2504.02495 Probably, the most popular study. It proposes to boost inference-time scalability by improving reward modeling. To enhance performance, DeepSeek-GRM uses adaptive critiques, parallel sampling, pointwise generative RM, and Self-Principled Critique Tuning (SPCT) 2. https://huggingface.co/papers/2504.04718 Allows small models to use external tools, like code interpreters and calculator, to enhance self-verification 3. https://huggingface.co/papers/2504.00810 Proposes to train LLMs on code-based reasoning paths to make test-time scaling more efficient, limiting unnecessary tokens with a special dataset and a Shifted Thinking Window 4. https://huggingface.co/papers/2504.00891 Introduces GenPRM, a generative PRM, that uses CoT reasoning and code verification for step-by-step judgment. With only 23K training examples, GenPRM outperforms prior PRMs and larger models 5. https://huggingface.co/papers/2503.24320 SWIFT test-time scaling framework improves World Models' performance without retraining, using strategies like fast tokenization, Top-K pruning, and efficient beam search 6. https://huggingface.co/papers/2504.07104 Proposes REBEL for RAG systems scaling, which uses multi-criteria optimization with CoT prompting for better performance-speed tradeoffs as inference compute increases 7. https://huggingface.co/papers/2503.13288 Proposes a φ-Decoding strategy that uses foresight sampling, clustering and adaptive pruning to estimate and select optimal reasoning steps Read further below 👇 Also, subscribe to the Turing Post https://www.turingpost.com/subscribe

replied to their post 3 days ago

16 new research on inference-time scaling: For the last couple of weeks a large amount of studies on inference-time scaling has emerged. And it's so cool, because each new paper adds a trick to the toolbox, making LLMs more capable without needing to scale parameter count of the models. So here are 13 new methods + 3 comprehensive studies on test-time scaling: 1. https://huggingface.co/papers/2504.02495 Probably, the most popular study. It proposes to boost inference-time scalability by improving reward modeling. To enhance performance, DeepSeek-GRM uses adaptive critiques, parallel sampling, pointwise generative RM, and Self-Principled Critique Tuning (SPCT) 2. https://huggingface.co/papers/2504.04718 Allows small models to use external tools, like code interpreters and calculator, to enhance self-verification 3. https://huggingface.co/papers/2504.00810 Proposes to train LLMs on code-based reasoning paths to make test-time scaling more efficient, limiting unnecessary tokens with a special dataset and a Shifted Thinking Window 4. https://huggingface.co/papers/2504.00891 Introduces GenPRM, a generative PRM, that uses CoT reasoning and code verification for step-by-step judgment. With only 23K training examples, GenPRM outperforms prior PRMs and larger models 5. https://huggingface.co/papers/2503.24320 SWIFT test-time scaling framework improves World Models' performance without retraining, using strategies like fast tokenization, Top-K pruning, and efficient beam search 6. https://huggingface.co/papers/2504.07104 Proposes REBEL for RAG systems scaling, which uses multi-criteria optimization with CoT prompting for better performance-speed tradeoffs as inference compute increases 7. https://huggingface.co/papers/2503.13288 Proposes a φ-Decoding strategy that uses foresight sampling, clustering and adaptive pruning to estimate and select optimal reasoning steps Read further below 👇 Also, subscribe to the Turing Post https://www.turingpost.com/subscribe

View all activity

Organizations

Kseniase's activity

published an article 11 days ago

Article

Topic 33: Slim Attention, KArAt, XAttention and Multi-Token Attention Explained – What’s Really Changing in Transformers?

By

and 1 other •

11 days ago

• 14

published an article 27 days ago

Article

What is Qwen-Agent framework? Inside the Qwen family

By

and 1 other •

27 days ago

• 8

published an article 29 days ago

Article

🌁#92: Fight for Developers and the Year of Orchestration

By

•

29 days ago

• 5

published an article 30 days ago

Article

🦸🏻#14: What Is MCP, and Why Is Everyone – Suddenly!– Talking About It?

By

•

30 days ago

• 165

published an article about 1 month ago

Article

How to Reduce Memory Use in Reasoning Models

By

and 1 other •

Mar 13

• 14

published an article about 1 month ago

Article

🌁#91: We are failing in AI literacy

By

and 1 other •

Mar 10

• 3

published an article about 1 month ago

Article

🌁#90: Why AI’s Reasoning Tests Keep Failing Us

By

•

Mar 3

• 9

published an article about 1 month ago

Article

🦸🏻#13: Action! How AI Agents Execute Tasks with UI and API Tools

By

•

Mar 10

• 8

published an article about 1 month ago

Article

🦸🏻#12: How Do Agents Learn from Their Own Mistakes? The Role of Reflection in AI

By

•

Mar 9

• 6

published an article about 1 month ago

Article

Everything You Need to Know about Knowledge Distillation

By

and 1 other •

Mar 6

• 22

published an article about 2 months ago

Article

Inside the family of Smol models

By

and 1 other •

Feb 27

• 10

published an article about 2 months ago

Article

🌁#89: AI in Action: How AI Engineers, Self-Optimizing Models, and Humanoid Robots Are Reshaping 2025

By

•

Feb 25

• 4

published an article about 2 months ago

Article

🦸🏻#11: How Do Agents Plan and Reason?

By

•

Feb 24

• 10

published an article about 2 months ago

Article

Topic 28: What is Mixture-of-Mamba?

By

and 1 other •

Feb 20

• 3

published an article about 2 months ago

Article

🌁#88: Can DeepSeek Inspire Global Collaboration?

By

•

Feb 17

• 3

published an article about 2 months ago

Article

🦸🏻#10: Does Present-Day GenAI Actually Reason?

By

•

Feb 15

• 7

published an article 2 months ago

Article

Topic 27: What are Chain-of-Agents and Chain-of-RAG?

By

and 1 other •

Feb 13

• 13

published an article 2 months ago

Article

🌁#87: Why DeepResearch Should Be Your New Hire

By

•

Feb 10

• 5