LLM Reasoning

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

68 papers in the last four weeks, up 143% on the four weeks before. 0.7% of all new papers.

Jul 13Week of Sep 28

Latest papers 654

All topics

Reasoning Can Be Restored by Correcting a Few Decision Tokens
May 16, 2026Changshuo Shen, Leheng Sheng, Yuxin Chen +2Efficient Language Model Reasoning Inference-Time Compute Allocation
University of Science and Technology of China, Hefei, China · National University of Singapore, Singapore
Confidence Geometry Reveals Trace-Level Correctness in Large Language Model Reasoning
May 16, 2026Shuo Liu, Ding Liu, Shi-Ju RanConfidence Estimation in Language Models LLM Reliability
School of Computer Science and Technology, Tiangong University, Tianjin, China. · Center for Quantum Physics and Intelligent Sciences, Department of Physics, Capital Normal University, Beijing, China.
Distributional Energy-Based Models for Uncertainty-Aware Structured LLM Reasoning
May 15, 2026Shireen Kudukkil Manchingal, Abhey Kalia, Fernanda Gonçalves +1LLM Answer Verification Energy-Based Models
Oxford Dynamics · Harwell Science and Innovation Campus, United Kingdom
SGR: A Stepwise Reasoning Framework for LLMs with External Subgraph Generation
May 15, 2026Xin Zhang, Yang Cao, Baoxing Wu +2LLM Grounding Knowledge Augmentation for Language Models
School of Information Science and Engineering, Chongqing Jiaotong University · School of Computer Science and Technology, Chongqing University of Posts and Telecommunications
MHGraphBench: Knowledge Graph-Grounded Benchmarking of Mental Health Knowledge in Large Language Models
May 15, 2026Weixin Liu, Congning Ni, Shelagh A. Mulvaney +4LLM Evaluation Healthcare
Vanderbilt University, Nashville, TN, USA · Vanderbilt University Medical Center, Nashville, TN, USA · Virginia Tech, Blacksburg, VA, USA
CAPS: Cascaded Adaptive Pairwise Selection for Efficient Parallel Reasoning
May 15, 2026Fangzhou Lin, Shuo Xing, Peiran Li +6Pairwise Comparison Pairwise Preference Evaluation
Texas A&M University · Georgia Institute of Technology · Tohoku University +1
Reasoning Models Don't Just Think Longer, They Move Differently
May 14, 2026Anders Gjølbye, Lars Kai Hansen, Sanmi KoyejoLLM Interpretability Representation Geometry in Language Models
Technical University of Denmark · Stanford University
OpenDeepThink: Parallel Reasoning via Bradley-Terry Aggregation
May 14, 2026Shang Zhou, Wenhao Chai, Kaiyuan Liu +3Pairwise Comparison Test-Time Optimization
1UC San Diego · 2Princeton University · University of Washington +1
A Deterministic Agentic Workflow for HS Tariff Classification: Multi-Dimensional Rule Reasoning with Interpretable Decisions
May 14, 2026Yu Zhang, Dongjiang Zhuang, Qu Zhou +4Agentic Workflows Classification
School of Information and Electronic Engineering, Shanghai Jiao Tong University, Shanghai, China · Department of Science and Technology (Shanghai), General Administration of Customs of the P.R.C. (GACC), Shanghai, China · Nanjing Jiyun Information Technology Co., Ltd., Nanjing, China +2
From Table to Cell: Attention for Better Reasoning with TABALIGN
May 14, 2026Tung Sum Thomas Kwok, Zeyong Zhang, Xinyu Wang +6Table QA Structured Reasoning
University of California, Los Angeles · SimpleWay · New Jersey Institute of Technology +4
Stateful Reasoning via Insight Replay
May 14, 2026Bin Lei, Caiwen Ding, Jiachen Yang +2Memory-Augmented Language Models CoT Reasoning
University of Minnesota · 2Simular AI
Darwin Family: MRI-Trust-Weighted Evolutionary Merging for Training-Free Scaling of Language-Model Reasoning
May 14, 2026Taebong Kim, Youngsik Hong, Minsik Kim +4Language Model Merging Model Merging
VIDRAFT Inc., Seoul, South Korea
Uncovering the Representation Geometry of Minimal Cores in Overcomplete Reasoning Traces
May 14, 2026Sanjoy Chowdhury, Dinesh ManochaNeural Representation Geometry LLM Interpretability
University of Maryland, College Park, USA
LLMs Know When They Know, but Do Not Act on It: A Metacognitive Harness for Test-time Scaling
May 13, 2026Qi Cao, Yufan Wang, Peijia Qin +2LLM Inference Test-Time Scaling
University of California, San Diego
Inference-Time Amplification of Weak Reasoning Models
May 13, 2026Varun Sunkaraneni, Pierfrancesco Beneventano, Riccardo Neumarker +2Inference-Time Scaling LLM Reasoning
MIT
Pause and Reflect: Conformal Aggregation for Chain-of-Thought Reasoning
May 13, 2026Yu Gu, Zijun Yu, Vahid Partovi Nia +1Selective Prediction Self-Consistency Decoding
Department of Mathematics and Statistics McGill University Montréal, QC, Canada · Department of Mathematics and Industrial Engineering Polytechnique de Montréal Montréal, QC, Canada
Enhanced and Efficient Reasoning in Large Learning Models
May 13, 2026Leslie G. ValiantRepresentation Learning World Models
John A. Paulson School of Engineering and Applied Sciences Harvard University, Boston, MA 02134
(How) Do Large Language Models Understand High-Level Message Sequence Charts?
May 13, 2026Mohammad Reza MousaviLLM Evaluation Chart Understanding
Department of Informatics, King’s College London
What properties of reasoning supervision are associated with improved downstream model quality?
May 13, 2026Mikołaj Langner, Dzmitry Pihulski, Jan Eliasz +5Training Data Selection LLM Reasoning
Wroclaw Tech, 50-370 Wrocław, Poland
Respecting Self-Uncertainty in On-Policy Self-Distillation for Efficient LLM Reasoning
May 13, 2026Junlong Ke, Zichen Wen, Weijia Li +2On-Policy Self-Distillation RL for Language Model Reasoning
2Tsinghua University · 1Shanghai Jiao Tong University · 3Shanghai AI Laboratory
GateKD: Confidence-Gated Closed-Loop Distillation for Robust Reasoning
May 13, 2026Kasidit Sermsri, Teerapong PanboonyuenCoT Distillation Language Model Distillation
†Chulalongkorn University · ‡MARSAIL · §PBYAIL (Panboonyuen AI Lab)
CommonWhy: A Dataset for Evaluating Entity-Based Causal Commonsense Reasoning in Large Language Models
May 13, 2026Armin Toroghi, Faeze Moradi Kalarde, Scott SannerAbductive Reasoning Causal Reasoning in Language Models
University of Toronto · Toronto, ON, Canada
LGMT: Logic-Grounded Metamorphic Testing for Evaluating the Reasoning Reliability of LLMs
May 12, 2026Zenghui Zhou, Man Li, Xiaoke Fang +3LLM Evaluation Automated Software Testing
School of Automation Science and Electrical Engineering, Beihang University, Xueyuan Road, Haidian District, Beijing, 100191, China
Solve the Loop: Attractor Models for Language and Reasoning
May 12, 2026Jacob Fein-Ashley, Paria RashidinejadLanguage Model Pretraining Recurrent Transformers
University of Southern California
OGLS-SD: On-Policy Self-Distillation with Outcome-Guided Logit Steering for LLM Reasoning
May 12, 2026Yuxiao Yang, Xiaoyun Wang, Weitong ZhangLanguage Model Calibration Language Model Distillation
UNC Chapel Hill · NVIDIA
Context Convergence Improves Answering Inferential Questions
May 12, 2026Jamshid Mozafari, Bhawna Piryani, Adam JatowtQuestion Answering LLM Reasoning
University of Innsbruck Innsbruck, Austria
Goal-Oriented Reasoning for RAG-based Memory in Conversational Agentic LLM Systems
May 12, 2026Jiazhou Liang, Armin Toroghi, Yifan Simon Liu +3Multi-Hop QA Agentic RAG
University of Toronto · Vector Institute for Artificial Intelligence
From Noise to Diversity: Random Embedding Injection in LLM Reasoning
May 12, 2026Heejun Kim, Seungpil Lee, Jewon Yeom +5LLM Prompting LLM Reasoning
1Gwangju Institute of Science and Technology · 3Microsoft Research · 2Seoul National University
Evolutionary Task Discovery: Advancing Reasoning Frontiers via Skill Composition and Complexity Scaling
May 12, 2026Liqin Ye, Yanbin Yin, Michael Galarnyk +3Evolutionary Optimization Curriculum Learning
Georgia Institute of Technology
Targeted Tests for LLM Reasoning: An Audit-Constrained Protocol
May 12, 2026Hongmin LiPrompt Sensitivity Language Model Robustness
School of Life Science and Technology, Institute of Science Tokyo · Department of Computational Biology and Medical Sciences, Graduate School of Frontier Sciences, The University of Tokyo

May 16, 2026, cs.AIJ/K move · Enter open · S save

Reasoning Can Be Restored by Correcting a Few Decision Tokens

Changshuo Shen, Leheng Sheng, Yuxin Chen, An Zhang+1

University of Science and Technology of China, Hefei, China · National University of Singapore, Singapore

Efficient Language Model Reasoning Inference-Time Compute Allocation LLM Routing LLM Reasoning Large Reasoning Models

Open paper AskPDF arXiv

Large reasoning models (LRMs) substantially outperform their base LLM counterparts on challenging reasoning benchmarks, yet it remains poorly understood where base models go wrong during token-by-token generation and how to narrow this gap efficiently. We study the base-reasoning gap through quantifying token-level distributional disagreement between a base model and a stronger reasoning model using likelihood-based divergences. Across benchmarks, we find that the reasoning advantage is highly sparse and concentrates on a small set of early, planning-related decision tokens. For instance, on Qwen3-0.6B, only ~8% of generated tokens account for the salient disagreement, and these tokens concentrate early in the response, are strongly enriched in planning-related decisions (17x), and coincide with high base-model uncertainty -- suggesting that base models fail mainly at early planning points that steer the subsequent reasoning trajectory. Building on these findings, we propose disagreement-guided token intervention, a simple inference-time delegation scheme that performs a one-token takeover by the reasoning model only at high-disagreement positions and immediately switches back to the base model. With a small intervention budget, this sparse delegation substantially recovers and can even surpass the performance of a same-size reasoning model on challenging reasoning tasks. Code is available at https://github.com/AlphaLab-USTC/RRTokenIntervention.

Momentum

Related topics

Latest papers 654