paper-with-me

Papers

Reasoning as State Transition: A Representational Analysis of Reasoning Evolution in Large Language Models

2026-01-31 · Siyuan Zhang, Jialian Li, Yichi Zhang, Xiao Yang, Yinpeng Dong, Hang Su arxiv

Large Language Models have achieved remarkable performance on reasoning tasks, motivating research into how this ability evolves during training. Prior work has primarily analyzed this evolution via explicit generation outcomes, treating the reasoning process as a black box and obscuring internal changes. To address this opacity, we introduce a representational perspective to investigate the dynamics of the model's internal states. Through comprehensive experiments across models at various training stages, we discover that post-training yields only limited improvement in static initial representation quality. Furthermore, we reveal that, distinct from non-reasoning tasks, reasoning involves a significant continuous distributional shift in representations during generation. Comparative analysis indicates that post-training empowers models to drive this transition toward a better distribution for task solving. To clarify the relationship between internal states and external outputs, statistical analysis confirms a high correlation between generation correctness and the final representations; while counterfactual experiments identify the semantics of the generated tokens, rather than additional computation during inference or intrinsic parameter differences, as the dominant driver of the transition. Collectively, we offer a novel understanding of the reasoning process and the effect of training on reasoning enhancement, providing valuable insights for future model analysis and optimization.

📄 PDF Abstract BibTeX arXiv:2602.00770

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

The Geometry of Thought: How Scale Restructures Reasoning In Large Language Models

2026-01-19 · Samuel Cyrenius Anderson arxiv

Scale does not uniformly improve reasoning - it restructures it. Analyzing 25,000+ chain-of-thought trajectories across four domains (Law, Science, Code, Math) and two scales (8B, 70B parameters), we discover that neural…

Mathematical ReasoningLegal Reasoning

The Reasoning Error About Reasoning: Why Different Types of Reasoning Require Different Representational Structures

2026-03-23 · Yiling Wu arxiv

Different types of reasoning impose different structural demands on representational systems, yet no systematic account of these demands exists across psychology, AI, and philosophy of mind. I propose a framework identif…

Causal InferenceFormal Logic

Explainable Chain-of-Thought Reasoning: An Empirical Analysis on State-Aware Reasoning Dynamics

2025-08-29 · Sheldon Yu, Yuxin Xiong, Junda Wu, Xintong Li 외 arxiv

Recent advances in chain-of-thought (CoT) prompting have enabled large language models (LLMs) to perform multi-step reasoning. However, the explainability of such reasoning remains limited, with prior work primarily focu…

Beyond Dense States: Elevating Sparse Transcoders to Active Operators for Latent Reasoning

2026-02-02 · Yadong Wang, Haodong Chen, Yu Tian, Chuanxing Geng 외 arxiv

Latent reasoning compresses the chain-of-thought (CoT) into continuous hidden states, yet existing methods rely on dense latent transitions that remain difficult to interpret and control. Meanwhile, sparse representation…

CTRLS: Chain-of-Thought Reasoning via Latent State-Transition

2025-07-10 · Junda Wu, Yuxin Xiong, Xintong Li, Zhengmian Hu 외

Chain-of-thought (CoT) reasoning enables large language models (LLMs) to break down complex problems into interpretable intermediate steps, significantly enhancing model transparency and performance in reasoning tasks. H…

Distributional Reinforcement Learningreinforcement-learningReinforcement Learning