English
Related papers

Related papers: Internal states before wait modulate reasoning pat…

200 papers

Language models often solve complex tasks by generating long reasoning chains, consisting of many steps with varying importance. While some steps are crucial for generating the final answer, others are removable. Determining which steps…

Computation and Language · Computer Science 2026-04-21 Yaniv Nikankin , Martin Tutek , Tomer Ashuach , Jonathan Rosenfeld , Yonatan Belinkov

Large language models excel on static benchmarks, but their ability as self-learning agents in dynamic environments remains unclear. We evaluate three prompting strategies: self-reflection, heuristic mutation, and planning across dynamic…

Artificial Intelligence · Computer Science 2025-08-12 Annie Wong , Thomas Bäck , Aske Plaat , Niki van Stein , Anna V. Kononova

Large language models (LLMs) can produce chains of thought (CoT) that do not accurately reflect the actual factors driving their answers. In multiple-choice settings with an injected hint favoring a particular option, models may shift their…

Machine Learning · Computer Science 2026-03-19 Parsa Mirtaheri , Mikhail Belkin

When a language model generates text, the selection of individual tokens might lead it down very different reasoning paths, making uncertainty difficult to quantify. In this work, we consider whether reasoning language models represent the…

Computation and Language · Computer Science 2025-11-07 Amir Zur , Atticus Geiger , Ekdeep Singh Lubana , Eric Bigelow

As chain-of-thought (CoT) has become central to scaling reasoning capabilities in large language models (LLMs), it has also emerged as a promising tool for interpretability, suggesting the opportunity to understand model decisions through…

Artificial Intelligence · Computer Science 2026-03-03 Kyle Cox , Darius Kianersi , Adrià Garriga-Alonso

Exploration capacity shapes both inference-time performance and reinforcement learning (RL) training for large (vision-) language models, as stochastic sampling often yields redundant reasoning paths with little high-level diversity. This…

Computer Vision and Pattern Recognition · Computer Science 2025-12-22 Rujiao Long , Yang Li , Xingyao Zhang , Weixun Wang , Tianqianjin Lin , Xi Zhao , Yuchi Xu , Wenbo Su , Junchi Yan , Bo Zheng

Large language models (LLMs) increasingly exhibit behaviors suggesting awareness of their evaluation context, often adapting their reasoning strategies in benchmark settings. Prior work has shown that such evaluation awareness can distort…

Computation and Language · Computer Science 2026-05-12 Yanshi Li , Xueru Bai , Shuman Liu , Haibo Zhang , Anxiang Zeng

In this work, we examine how targeted perturbations in the activation space of Language Models (LMs) can encode complex reasoning patterns. We inject steering vectors, derived from LM activations, into LMs during inference time and study…

Computation and Language · Computer Science 2025-03-24 Jason Zhang , Scott Viteri

Large language models (LLMs) can perform reasoning computations both internally within their latent space and externally by generating explicit token sequences like chains of thought. Significant progress in enhancing reasoning abilities…

Computation and Language · Computer Science 2025-04-16 Thilo Hagendorff , Sarah Fabi

Chain of Thought (CoT) prompting can encourage language models to engage in multi-step logical reasoning. The quality of the provided demonstrations significantly influences the success of downstream inference tasks. Current unsupervised…

Computation and Language · Computer Science 2025-05-27 Yufeng Zhang , Xuepeng Wang , Lingxiang Wu , Jinqiao Wang

While reasoning models are increasingly ubiquitous, the effects of reasoning training on a model's internal mechanisms remain poorly understood. In this work, we introduce transcoder adapters, a technique for learning an interpretable…

Machine Learning · Computer Science 2026-02-25 Nathan Hu , Jake Ward , Thomas Icard , Christopher Potts

Chain-of-Thought reasoning has emerged as a pivotal methodology for enhancing model inference capabilities. Despite growing interest in Chain-of-Thought reasoning, its underlying mechanisms remain unclear. This paper explores the working…

Computation and Language · Computer Science 2025-09-03 Hao Yang , Zhiyu Yang , Yunjie Zhang , Shanyi Zhu , Lin Yang

Latent tokens are gaining attention for enhancing reasoning in large language models (LLMs), yet their internal mechanisms remain unclear. This paper examines the problem from a reliability perspective, uncovering fundamental weaknesses:…

Computation and Language · Computer Science 2025-12-29 Yuyi Zhang , Boyu Tang , Tianjie Ju , Sufeng Duan , Gongshen Liu

Thinking LLMs produce reasoning traces before answering. Prior activation steering work mainly targets on shaping these traces. It remains less understood how answer tokens actually read and integrate the reasoning to produce reliable…

Computation and Language · Computer Science 2026-04-22 Haoyang Chen , Yi Liu , Jianzhi Shao , Tao Zhang , Chengfu Huo , Wei Hu

Reasoning-enhanced large language models (LLMs) explicitly generate intermediate reasoning steps prior to generating final answers, helping the model excel in complex problem-solving. In this paper, we demonstrate that this emerging…

Machine Learning · Computer Science 2025-05-22 Tong Wu , Chong Xiang , Jiachen T. Wang , G. Edward Suh , Prateek Mittal

Test-time inference has emerged as a powerful paradigm for enabling language models to ``think'' longer and more carefully about complex challenges, much like skilled human experts. While reinforcement learning (RL) can drive…

Computation and Language · Computer Science 2025-08-18 Kanishk Gandhi , Ayush Chakravarthy , Anikait Singh , Nathan Lile , Noah D. Goodman

The chain-of-thought (CoT) paradigm uses the elicitation of step-by-step rationales as a proxy for reasoning, gradually refining the model's latent representation of a solution. However, it remains unclear just how early a Large Language…

Computation and Language · Computer Science 2025-11-20 Joey David

Large reasoning models (LRMs) achieve strong performance on mathematical reasoning tasks, often attributed to their capability to generate explicit chain-of-thought (CoT) explanations. However, recent work shows that LRMs often arrive at…

Computation and Language · Computer Science 2026-04-20 Yihong Liu , Raoyuan Zhao , Hinrich Schütze , Michael A. Hedderich

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…

Computation and Language · Computer Science 2026-02-03 Siyuan Zhang , Jialian Li , Yichi Zhang , Xiao Yang , Yinpeng Dong , Hang Su

The recent advent of reasoning models like OpenAI's o1 was met with excited speculation by the AI community about the mechanisms underlying these capabilities in closed models, followed by a rush of replication efforts, particularly from…

Computation and Language · Computer Science 2025-11-21 Brown Ebouky , Andrea Bartezzaghi , Mattia Rigotti