paper-with-me

Papers

Demystifying Long Chain-of-Thought Reasoning in LLMs

2025-02-05 · Edward Yeo, Yuxuan Tong, Morry Niu, Graham Neubig, Xiang Yue

Scaling inference compute enhances reasoning in large language models (LLMs), with long chains-of-thought (CoTs) enabling strategies like backtracking and error correction. Reinforcement learning (RL) has emerged as a crucial method for developing these capabilities, yet the conditions under which long CoTs emerge remain unclear, and RL training requires careful design choices. In this study, we systematically investigate the mechanics of long CoT reasoning, identifying the key factors that enable models to generate long CoT trajectories. Through extensive supervised fine-tuning (SFT) and RL experiments, we present four main findings: (1) While SFT is not strictly necessary, it simplifies training and improves efficiency; (2) Reasoning capabilities tend to emerge with increased training compute, but their development is not guaranteed, making reward shaping crucial for stabilizing CoT length growth; (3) Scaling verifiable reward signals is critical for RL. We find that leveraging noisy, web-extracted solutions with filtering mechanisms shows strong potential, particularly for out-of-distribution (OOD) tasks such as STEM reasoning; and (4) Core abilities like error correction are inherently present in base models, but incentivizing these skills effectively for complex tasks via RL demands significant compute, and measuring their emergence requires a nuanced approach. These insights provide practical guidance for optimizing training strategies to enhance long CoT reasoning in LLMs. Our code is available at: https://github.com/eddycmu/demystify-long-cot.

📄 PDF Abstract BibTeX arXiv:2502.03373

Code (1)

eddycmu/demystify-long-cot 공식 구현 pytorch

Tasks

Reinforcement Learning (RL)

Methods 이 논문이 사용한 방법론

BASE 설명 없음
SFT Shrink and Fine-Tune, or SFT, is a type of distillation that avoids explicit distillation by copying parameters to a student student model and then fine-tuning.…

Similar Papers 제목 키워드 기반

Demystifying Entropy-based Selection for Chain-of-Thought Compression in Large Reasoning Models

2026-07-30 · Sara Candussio, Daniel Scalena, Luca Bortolussi, Elisabetta Fersini 외 arxiv

Entropy-based pruning has been proposed as an effective method for compressing Chain-of-Thought (CoT) reasoning with negligible accuracy loss. We test the robustness of low- and high-entropy CoT step selection methods ac…

Demystifying Multilingual Chain-of-Thought in Process Reward Modeling

2025-02-18 · Weixuan Wang, Minghao Wu, Barry Haddow, Alexandra Birch

Large language models (LLMs) are designed to perform a wide range of tasks. To improve their ability to solve complex problems requiring multi-step reasoning, recent research leverages process reward modeling to provide …

Reinforcement Learning (RL)

Deconstructing Long Chain-of-Thought: A Structured Reasoning Optimization Framework for Long CoT Distillation

2025-03-20 · Yijia Luo, Yulin Song, Xingyao Zhang, Jiaheng Liu 외

Recent advancements in large language models (LLMs) have demonstrated remarkable reasoning capabilities through long chain-of-thought (CoT) reasoning. The R1 distillation scheme has emerged as a promising approach for tr…

Unlocking General Long Chain-of-Thought Reasoning Capabilities of Large Language Models via Representation Engineering

2025-03-14 · Xinyu Tang, Xiaolei Wang, Zhihao Lv, Yingqian Min 외

Recent advancements in long chain-of-thoughts(long CoTs) have significantly improved the reasoning capabilities of large language models(LLMs). Existing work finds that the capability of long CoT reasoning can be efficie…

Hierarchical Chain-of-Thought: Enhancing LLM Reasoning Performance and Efficiency

2026-03-31 · Xingshuai Huang, Derek Li, Bahareh Nikpour, Parsa Omidi arxiv

Chain-of-Thought (CoT) prompting has significantly improved the reasoning capabilities of large language models (LLMs). However, conventional CoT often relies on unstructured, flat reasoning chains that suffer from redun…

Mathematical Reasoning