LLM Self-Correction

LLM: Large Language Model

This field asks whether language models can recognize faulty claims, reasoning steps, or actions and revise them, which matters for reliability in tasks such as mathematics, coding, question answering, and clinical workflows. Approaches include prompted reflection and critique, feedback from code execution or formal verifiers, retrieval and multi-agent review, training with reinforcement learning on failed attempts, and signals from model uncertainty or internal representations. Evaluations use settings such as GSM8K and AIME for mathematics, HumanEval and SWE-bench for code, and Text-to-SQL benchmarks, often tracking both errors fixed and correct answers damaged. A central tension is that detecting an error does not ensure a useful repair: noisy feedback, poor error localization, or repeated revision can worsen outputs, motivating selective, evidence-grounded correction and reliable stopping rules.

Latest papers

159 papers24 papers in the last four weeks, up 167% on the four weeks before. 0.1% of all new papers.

Remask, Don't Replace: Token-to-Mask Refinement in Diffusion Large Language Models
Apr 20, 2026Lin YaoLLM Self-Correction Multi-Token Prediction
School of Computer Science, Shanghai Jiao Tong University, Shanghai, 200240, China · Zhongguancun Academy, Beijing, 100097, China
Latent Phase-Shift Rollback: Inference-Time Error Correction via Residual Stream Monitoring and KV-Cache Steering
Apr 20, 2026Manan Gupta, Dhruv KumarLLM Self-Correction Language Model Error Detection
1BITS Pilani, Pilani Campus, India
SPREG: Structured Plan Repair with Entropy-Guided Test-Time Intervention for Large Language Model Reasoning
Apr 20, 2026Xuan Wang, Yu Ming, Xinhao Zhong +4LLM Self-Correction Classifier-Free Guidance
1Meituan
Improving Retrieval-Augmented Generation without Taxonomy-based Error Categorization
Apr 16, 2026Gongbo Zhang, Yifan Peng, Chunhua WengLLM Self-Correction Retrieval-Augmented Generation
Columbia University · Weill Cornell Medicine
Vibe-Coding: Feedback-Based Automated Verification with no Human Code Inspection, a Feasibility Study
Apr 16, 2026Michal Töpfer, František Plášil, Tomáš Bureš +1LLM Self-Correction Formal Verification
Charles University Prague, Czech Republic
Enhancing Reinforcement Learning for Radiology Report Generation with Evidence-aware Rewards and Self-correcting Preference Learning
Apr 15, 2026Qin Zhou, Guoyan Liang, Qianyi Yang +4Radiology Report Generation RL for Generative Models
Department of Computer Science and Engineering, ECUST, China · Key Laboratory of Smart Manufacturing in Energy Chemical Process, Ministry of Education, P. R. China · Zhejiang University, Hangzhou, China +1
SQL Query Engine: A Self-Healing LLM Pipeline for Natural Language to PostgreSQL Translation
Apr 15, 2026Muhammad Adeel IjazExecution-Guided Text-to-SQL Text-to-SQL
REGREACT: Self-Correcting Multi-Agent Pipelines for Structured Regulatory Information Extraction
Apr 13, 2026Mohammed Ali, Abdelrahman Abdallah, Adam JatowtLLM Self-Correction Document Information Extraction
University of Innsbruck, Austria
How Many Tries Does It Take? Iterative Self-Repair in LLM Code Generation Across Model Scales and Benchmarks
Apr 12, 2026Johin Johny ArimburLLM Self-Correction LLM Code Generation
Independent Researcher
Agentic Application in Power Grid Static Analysis: Automatic Code Generation and Error Correction
Apr 11, 2026Qinjuan Wang, Shan Yang, Yongli ZhuAgentic Code Generation LLM-Based Program Synthesis
Sun Yat-sen University, Guangzhou, China
Constraint-Aware Corrective Memory for Language-Based Drug Discovery Agents
Apr 10, 2026Maochen Sun, Youzhi Zhang, Gaofeng MengDrug Discovery Memory-Augmented Language Model Agents
Institute of Automation, Chinese Academy of Sciences, China · School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences, China · Centre for Artificial Intelligence and Robotics, Hong Kong Institute of Science & Innovation, Chinese Academy of Sciences, China
Self-Debias: Self-correcting for Debiasing Large Language Models
Apr 9, 2026Xuan Feng, Shuai Zhao, Luwei Xiao +2Social Bias in Language Models LLM Self-Correction
Jinan University, China · Nanyang Technological University, Singapore
From Hallucination to Structure Snowballing: The Alignment Tax of Constrained Decoding in LLM Reflection
Apr 7, 2026Hongxu ZhouSelf-Reflection in Language Model Agents LLM Self-Correction
Erasmus Mundus Master’s in Language and Communication Technologies Saarland University Saarbrücken, Germany
Benchmarking Multi-turn Medical Diagnosis: Hold, Lure, and Self-Correction
Apr 6, 2026Jinrui Fang, Runhan Chen, Xu Yang +9Clinical Language Model Evaluation Medical Diagnosis
The University of Texas at Austin · New York University · The University of Texas Southwestern Medical Center +2
NeuReasoner: Towards Explainable, Controllable, and Unified Reasoning via Mixture-of-Neurons
Apr 3, 2026Haonan Dong, Kehan Jiang, Haoran Ye +3Large Reasoning Models Efficient Language Model Reasoning
State Key Laboratory of General Artificial Intelligence, School of Intelligence Science and Technology, Peking University · School of Software and Microelectronics, Peking University
MM-ReCoder: Advancing Chart-to-Code Generation with Reinforcement Learning and Self-Correction
Apr 2, 2026Zitian Tang, Xu Zhang, Jianbo Yuan +4Chart-to-Code Generation Code Generation
Brown University · Amazon AGI
Can Large Language Models Self-Correct in Medical Question Answering? An Exploratory Study
Mar 31, 2026Zaifu Zhan, Mengyuan Cui, Rui ZhangLLM Self-Correction LLM Reliability
Department of Electrical and Computer Engineering, University of Minnesota, Minneapolis, USA · Division of Computational Health Sciences, Department of Surgery, University of Minnesota, Minneapolis, USA · Department of Software Engineering, University of St. Thomas, Saint Paul,USA
Silent Commitment Failure in Instruction-Tuned Language Models: Evidence of Governability Divergence Across Architectures
Mar 22, 2026Gregory M. RuddellLanguage Model Error Detection LLM Reliability
SnailSafe.ai — Carson City, Nevada
A Training-Free Regeneration Paradigm: Contrastive Reflection Memory Guided Self-Verification and Self-Improvement
Mar 20, 2026Yuran Li, Di Wu, Benoit BouletLanguage Model Self-Verification Language Model Self-Improvement
Department of Electrical and Computer Engineering, McGill University · Department of Mechanical, Industrial and Aerospace Engineering, Concordia University
Self-Aware Markov Models for Discrete Reasoning
Mar 17, 2026Gregor Kornhardt, Jannis Chemseddine, Christian Wald +1Masked Diffusion Models Adaptive Computation
Technische Universität Berlin, Straße des 17. Juni 136, 10623 Berlin, Germany
Decoding the Critique Mechanism in Large Reasoning Models
Mar 17, 2026Hoang Phan, Quang H. Nguyen, Hung T. Q. Le +3Language Model Self-Verification Large Reasoning Models
VinUni-Illinois Smart Health Center, VinUniversity · University of Illinois Urbana-Champaign
CounterRefine: Answer-Conditioned Counterevidence Retrieval for Inference-Time Knowledge Repair in Factual Question Answering
Mar 17, 2026Tianyi Huang, Ying Kai DengFactual QA Question Answering
Ryquo · App-In Club
VIBEPASS: Can Vibe Coders Really Pass the Vibe Check?
Mar 16, 2026Srijan Bansal, Jiao Fangkai, Yilun Zhou +3LLM Self-Correction Language Model Error Detection
Salesforce AI Research · Nanyang Technological University · A*STAR
Understanding Reasoning in LLMs through Strategic Information Allocation under Uncertainty
Mar 16, 2026Jeonghye Kim, Xufang Luo, Minbeom Kim +3LLM Reasoning LLM Self-Correction
Microsoft Research · KAIST · Seoul National University
Cross-Context Review: Improving LLM Output Quality by Separating Production and Review Sessions
Mar 12, 2026Tae-Eun SongLLM Self-Correction LLM Evaluation
Daejeon Jungang Cheonggua Co., Ltd.
AI Knows What's Wrong But Cannot Fix It: Helicoid Dynamics in Frontier LLMs Under High-Stakes Decisions
Mar 12, 2026Alejandro R JadadLLM Reliability LLM Decision-Making
Research Professor (Adjunct), Department of Population and Public Health Sciences, Keck School of Medicine, University of Southern California · Principal, Vivenxia Group, Los Angeles, California, USA
Self-Execution Simulation Improves Coding Models
Mar 11, 2026Gallil Maimon, Ori Yoran, Felix Kreuk +4Execution-Guided Code Generation LLM Self-Correction
FAIR team, Meta · Hebrew University of Jerusalem · Inria
Not All Queries Need Deep Thought: CoFiCot for Adaptive Coarse-to-fine Stateful Refinement
Mar 9, 2026Dongxu Zhang, Hongqiang Lin, Yiding Sun +4LLM Self-Refinement Efficient Language Model Reasoning
Xi’an Jiaotong University · Zhejiang University · The Chinese University of Hong Kong (Shenzhen) +1
ReflexiCoder: Teaching Large Language Models to Self-Reflect on Generated Code and Self-Correct It via Reinforcement Learning
Mar 6, 2026Juyong Jiang, Jiasi Shen, Sunghun Kim +3LLM Code Generation LLM Self-Correction
The Hong Kong University of Science and Technology (Guangzhou) · The Hong Kong University of Science and Technology · Amazon AGI +1
ErrorLLM: Modeling SQL Errors for Text-to-SQL Refinement
Mar 4, 2026Zijin Hong, Hao Chen, Zheng Yuan +6LLM Self-Correction Text-to-SQL
The Hong Kong Polytechnic University Kowloon, Hong Kong · City University of Macau Taipa, Macau · Jilin University Changchun, China +3

Apr 20, 2026J/K move · Enter open · S save

Remask, Don't Replace: Token-to-Mask Refinement in Diffusion Large Language Models

Lin Yao

School of Computer Science, Shanghai Jiao Tong University, Shanghai, 200240, China · Zhongguancun Academy, Beijing, 100097, China

LLM Self-Correction Multi-Token Prediction LLM Reliability Diffusion Language Model Decoding Diffusion Language Models

Open paper AskPDF arXiv

Diffusion large language models (dLLMs) gain speed by committing multiple tokens in parallel at each denoising step, but any erroneous commitment persists as conditioning context and biases every subsequent prediction. LLaDA2.1 repairs such errors with Token-to-Token (T2T) editing, which re-examines previously unmasked tokens and overwrites them when an alternative becomes sufficiently confident. We argue that this replacement action is itself the limiting factor: under polluted context, a confident replacement can propagate the error, while under a multimodal posterior no alternative may be confident enough to trigger an edit. We propose Token-to-Mask (T2M) remasking, a training-free rule that revokes suspicious commitments by resetting them to [M] and lets the subsequent mask-filling steps re-predict them from a cleaner context. T2M improves accuracy by +13.33 points on AIME 2025 and +8.56 points on CMATH. These results suggest that, for parallel discrete generators, remasking suspect tokens rather than overwriting them is a more reliable self-correction primitive.