LLM Self-Correction

LLM: Large Language Model

Overview

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

CardsList

171 papers13 papers in the last four weeks, down 46% on the four weeks before. 0.1% of all new papers.

  1. Found, but Not Fixed : From Software Vulnerability Discovery to Fixes with Iterative LLM Pipelines

    Nov 15, 2026Youssef Gamal, Mohamad Fakih, Halima Bouzidi +5Automated Program RepairLLM Self-Correction

  2. Agent-Guided Process Verification for Mathematical Self-Correction

    Nov 14, 2026Rui Zou, Mengqi Wei, Xin Zhao +1LLM Self-CorrectionMathematical Reasoning

  3. Catch It Where It Happens: Real Time Error Correction During LLM Reasoning

    Nov 9, 2026Rotem Elimelech, Michael Shalyt, Ido KaminerLanguage Model Error DetectionLLM Self-Correction

  4. BiVerify: Closing the Loop Between Backward and Forward Verification for Mathematical Self-Correction

    Nov 8, 2026Rui Zou, Xin Zhao, Mengqi Wei +1LLM Self-CorrectionLanguage Model Self-Verification

  5. On the Limits of Execution Traces for Repairing Assertion Failures in LLM-Generated Code

    Nov 6, 2026Muntasir Adnan, Carlos C. N. KuhnAutomated Program RepairLLM Self-Correction

  6. More with Less: Efficient Parallel Scientific Code Translation via an Agent-Skill Workflow

    Nov 6, 2026Jianuo Shen, Pengyi Xu, Bo ZhangCode TranslationLLM Agent Workflow Optimization

  7. When Self-Verification Hurts: Calibrated Answer Repair for Financial Question Answering

    Nov 3, 2026Celine Guo, Weida Liang, Yujian Fu +4Question AnsweringFinancial QA

  8. Toward Executable Multi-framework Front-end Code Generation with Self-Correction

    Nov 1, 2026Linxiao Li, Haoran Ma, Chenyue Wang +3Code GenerationImage-to-Code Generation

  9. Don't Forget What You Just Learned: Self-Directed Repair Turns Agent Mistakes Into Lasting Lessons

    Oct 22, 2026Yuval Kansal, Karen Ullrich, Polina Kirichenko +4Self-Improving AgentsLLM Self-Correction

  10. Text-Only LLM Correctors Break Correct Transcripts: A Repair-Damage Analysis of Audio Grounding in ASR Error Correction

    Oct 21, 2026Vasiliy Kudryavtsev, Kirill Borodin, Nikita Vasiliev +2LLM Self-CorrectionAutomatic Speech Recognition

  11. CorrectRTL: Boost RTL Generation Quality Through Automated Self-Correction with Instruction Decomposition

    Oct 18, 2026Jing Wang, Shang Liu, Hangan Zhou +4RTL Code GenerationLLM Self-Correction

  12. Repairing LLM-Generated Specifications with Deterministic Diagnosis

    Oct 15, 2026Giovanni Vaccarino, Emanuele La Malfa, Dehai Min +6AutoformalizationLLM Self-Correction

  13. Recovering Reliable Supervision via Critique-Guided On-Policy Distillation

    Oct 14, 2026Kaixiang Zhao, Tianrun Yu, Shangzhe Li +4On-Policy DistillationKnowledge Distillation

  14. ReFormBench: Benchmarking the Stability of Response-Conditioned Format Correction in Large Language Models

    Oct 14, 2026Tianxiang Gao, Yahan Li, Chenlu Guo +3LLM EvaluationInstruction Following

  15. REWIND: Uncertainty-Guided Trajectory Repair for Test-Time Reasoning

    Oct 14, 2026Yi Ding, Mengyu Wang, Min Zhou +1LLM ReasoningLLM Self-Correction

  16. Beyond Answer Voting: Fine-Grained Correction of LLM Reasoning through Graph Alignment across Sampled Traces

    Oct 13, 2026Haotian Liu, Yingjian Chen, Xiaohang Xu +4LLM Self-CorrectionReasoning Verification

  17. Reliable Text-to-SQL in Resource-Constrained Settings with Small LLMs via Iterative Retry Workflows

    Oct 12, 2026Bilel Abderrahmane Benziane, Xavier ScheplerText-to-SQLSmall Language Models

  18. Where Did We Go Wrong? Localizing Errors to Repair Multi-Turn Conversations

    Oct 10, 2026Alessandro Morosini, Wei Sun, Georgia PerakisState TrackingLanguage Model Error Detection

  19. Inject a Turn, Not a Token: Learning Where to Redirect LLM Reasoning for Self-Correction

    Oct 9, 2026Jiwon Jeon, Myungsik Cho, Jeonghye Kim +1LLM Self-CorrectionLLM Reasoning

  20. ReTeach: Building a Self-Teacher through Multi-Round Reflection and Retry

    Oct 8, 2026Yafeng Tang, Hao Li, Hongsheng Yu +1Language Model DistillationLLM Self-Refinement

  21. How Much Evidence Should a Coding Agent's Self-Correction Carry? Adaptive Dirichlet Evidence for Self-Distillation

    Oct 6, 2026Yunbo Long, Guangya Hao, Yuhan Liu +4LLM Self-CorrectionCoding Agents

  22. Can Language Models Learn to Reject Their Own Bad Reasoning Steps?

    Oct 5, 2026Siheng Xiong, Xiaoze Liu, Yiqiao Jin +2LLM Self-CorrectionLanguage Model Decoding