mlsanityData to 14 Oct 2026

Automatically Correcting Large Language Models: Surveying the landscape of diverse self-correction strategies

arXiv 2308.03188 ↗First posted 6 Aug 2023cs.CL cs.AI cs.LGcited by 238

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Large language models (LLMs) have demonstrated remarkable performance across a wide array of NLP tasks. However, their efficacy is undermined by undesired and inconsistent behaviors, including hallucination, unfaithful reasoning, and toxic content. A promising approach to rectify these flaws is self-correction, where the LLM itself is prompted or guided to fix problems in its own output. Techniques leveraging automated feedback -- either produced by the LLM itself or some external system -- are of particular interest as they are a promising way to make LLM-based solutions more practical and deployable with minimal human feedback. This paper presents a comprehensive review of this emerging class of techniques. We analyze and taxonomize a wide array of recent work utilizing these strategies, including training-time, generation-time, and post-hoc correction. We also summarize the major applications of this strategy and conclude by discussing future directions and challenges.

Key facts. Cited by 238 papers in the map (data through 14 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 137 of its 158 references that are in the map. First posted on arXiv 6 Aug 2023.

Builds on

137 of 158 references.

Closest ancestors · 9

Same direction, newest first

  1. 2023SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoning1 Aug
  2. 2023CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing19 May
  3. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Models17 May
  4. 2023REFINER: Reasoning Feedback on Intermediate Representations4 Apr
  5. 2023Self-Refine: Iterative Refinement with Self-Feedback30 Mar
  6. 2022Large Language Models Can Self-Improve20 Oct
  7. 2022Language Models (Mostly) Know What They Know11 Jul
  8. 2022Towards Teachable Reasoning Systems: Using a Dynamic Memory of User Feedback for Continual System Improvement27 Apr
  9. 2022STaR: Bootstrapping Reasoning With Reasoning28 Mar

Other references · 128

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2017Proximal Policy Optimization Algorithmscited by 15,215
  4. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  5. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  6. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  7. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  8. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  9. 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
  10. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
  11. 2023Let's Verify Step by Stepcited by 3,789
  12. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
38 more
  1. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  2. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
  3. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,975
  4. 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
  5. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
  6. 2022Emergent Abilities of Large Language Modelscited by 2,512
  7. 2020DeBERTa: Decoding-enhanced BERT with Disentangled Attentioncited by 2,181
  8. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,807
  9. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  10. 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,282
  11. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,255
  12. 2020BLEURT: Learning Robust Metrics for Text Generationcited by 1,203
  13. 2022MaskGIT: Masked Generative Image Transformercited by 1,099
  14. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,027
  15. 2023Teaching Large Language Models to Self-Debugcited by 988
  16. 2022Solving math word problems with process- and outcome-based feedbackcited by 901
  17. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 872
  18. 2023Reasoning with Language Model is Planning with World Modelcited by 861
  19. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 810
  20. 2019Plug and Play Language Models: A Simple Approach to Controlled Text Generationcited by 799
  21. 2019CTRL: A Conditional Transformer Language Model for Controllable Generationcited by 762
  22. 2023AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedbackcited by 748
  23. 2021FinQA: A Dataset of Numerical Reasoning over Financial Datacited by 562
  24. 2021Editing Factual Knowledge in Language Modelscited by 553
  25. 2022Improving alignment of dialogue agents via targeted human judgementscited by 531
  26. 2023DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Modelscited by 518
  27. 2023Is ChatGPT a General-Purpose Natural Language Processing Task Solver?cited by 488
  28. 2023How Close is ChatGPT to Human Experts? Comparison Corpus, Evaluation, and Detectioncited by 481
  29. 2023Fine-Grained Human Feedback Gives Better Rewards for Language Model Trainingcited by 401
  30. 2023Logic-LM: Empowering Large Language Models with Symbolic Solvers for Faithful Logical Reasoningcited by 391
  31. 2022CodeT: Code Generation with Generated Testscited by 384
  32. 2022Self-critiquing models for assisting human evaluatorscited by 380
  33. 2021FUDGE: Controlled Text Generation With Future Discriminatorscited by 376
  34. 2023Language Models can Solve Computer Taskscited by 372
  35. 2023Editing Large Language Models: Problems, Methods, and Opportunitiescited by 372
  36. 2022CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learningcited by 358
  37. 2023Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedbackcited by 353
  38. 2023How Language Model Hallucinations Can Snowballcited by 311

50 most cited of 128 All 128 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 33; 2024: 120; 2025: 63; 2026: 22.

238 papers.

Sort:

By area:

  1. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,601
  2. 2023Large Language Models Cannot Self-Correct Reasoning Yetcited by 1,017
  3. 2024Self-Rewarding Language Modelscited by 681
  4. 2024Training Language Models to Self-Correct via Reinforcement Learningcited by 426
  5. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Modelscited by 384
  6. 2024Hallucination is Inevitable: An Innate Limitation of Large Language Modelscited by 332
  7. 2024Deliberative Alignment: Reasoning Enables Safer Language Modelscited by 308
  8. 2023Large Language Model Alignment: A Surveycited by 243
  9. 2023PromptAgent: Strategic Planning with Language Models Enables Expert-level Prompt Optimizationcited by 233
  10. 2023A Survey of Large Language Models for Healthcare: from Data, Technology, and Applications to Accountability and Ethicscited by 163
  11. 2024Fine-Tuning Large Vision-Language Models as Decision-Making Agents via Reinforcement Learningcited by 158
  12. 2024LLM Critics Help Catch LLM Bugscited by 147
38 more
  1. 2023GPT-4V in Wonderland: Large Multimodal Models for Zero-Shot Smartphone GUI Navigationcited by 137
  2. 2024From Decoding to Meta-Generation: Inference-time Algorithms for Large Language Modelscited by 137
  3. 2023AutoMix: Automatically Mixing Language Modelscited by 125
  4. 2023Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselvescited by 123
  5. 2023AgentCF: Collaborative Learning with Autonomous Language Agents for Recommender Systemscited by 116
  6. 2024Pride and Prejudice: LLM Amplifies Self-Bias in Self-Refinementcited by 114
  7. 2023Self-correcting LLM-controlled Diffusion Modelscited by 113
  8. 2024Rethinking Interpretability in the Era of Large Language Modelscited by 111
  9. 2024Making Reasoning Matter: Measuring and Improving Faithfulness of Chain-of-Thought Reasoningcited by 110
  10. 2023Let's reward step by step: Step-Level reward model as the Navigators for Reasoningcited by 102
  11. 2023Volcano: Mitigating Multimodal Hallucination through Self-Feedback Guided Revisioncited by 98
  12. 2024Agent-Pro: Learning to Evolve via Policy-Level Reflection and Optimizationcited by 94
  13. 2024VideoHallucer: Evaluating Intrinsic and Extrinsic Hallucinations in Large Video-Language Modelscited by 90
  14. 2024Small Language Models Need Strong Verifiers to Self-Correct Reasoningcited by 86
  15. 2023LLMs cannot find reasoning errors, but can correct them given the error locationcited by 85
  16. 2024Long Is More for Alignment: A Simple but Tough-to-Beat Baseline for Instruction Fine-Tuningcited by 85
  17. 2023Cognitive Mirage: A Review of Hallucinations in Large Language Modelscited by 84
  18. 2024Formal Mathematical Reasoning: A New Frontier in AIcited by 81
  19. 2023Igniting Language Intelligence: The Hitchhiker's Guide From Chain-of-Thought Reasoning to Language Agentscited by 72
  20. 2024Self-Contrast: Better Reflection Through Inconsistent Solving Perspectivescited by 71
  21. 2025Self-rewarding correction for mathematical reasoningcited by 70
  22. 2024Can LLMs Learn from Previous Mistakes? Investigating LLMs' Errors to Boost for Reasoningcited by 69
  23. 2023Clover: Closed-Loop Verifiable Code Generationcited by 67
  24. 2024A Survey on Employing Large Language Models for Text-to-SQL Taskscited by 67
  25. 2024The Benefits of a Concise Chain of Thought on Problem-Solving in Large Language Modelscited by 63
  26. 2024Can Language Models Learn to Skip Steps?cited by 59
  27. 2024Selective Reflection-Tuning: Student-Selected Data Recycling for LLM Instruction-Tuningcited by 57
  28. 2024Multi-step Problem Solving Through a Verifier: An Empirical Analysis on Model-induced Process Supervisioncited by 52
  29. 2024Do LLMs Know about Hallucination? An Empirical Investigation of LLM's Hidden Statescited by 52
  30. 2023Towards Reasoning in Large Language Models via Multi-Agent Peer Review Collaborationcited by 51
  31. 2024DALK: Dynamic Co-Augmentation of LLMs and KG to answer Alzheimer's Disease Questions with Scientific Literaturecited by 50
  32. 2024Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problemscited by 50
  33. 2024Scaling of Search and Learning: A Roadmap to Reproduce o1 from Reinforcement Learning Perspectivecited by 50
  34. 2025DeepAnalyze: Agentic Large Language Models for Autonomous Data Sciencecited by 50
  35. 2024ALI-Agent: Assessing LLMs' Alignment with Human Values via Agent-based Evaluationcited by 47
  36. 2024Puzzle Solving using Reasoning of Large Language Models: A Surveycited by 45
  37. 2024In-Context Sharpness as Alerts: An Inner Representation Perspective for Hallucination Mitigationcited by 45
  38. 2024A Theoretical Understanding of Self-Correction through In-context Alignmentcited by 44

50 most cited of 238 All 238 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2024Large Language Models have Intrinsic Self-Correction Ability
  2. 2025Self-Improvement in Multimodal Large Language Models: A Survey
  3. 2026Self-Improvement of Large Language Models: A Technical Overview and Future Outlook
  4. 2025Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
  5. 2025Self-Correcting Large Language Models: Generation vs. Multiple Choice
  6. 2025Self-Correction Bench: Uncovering and Addressing the Self-Correction Blind Spot in Large Language Models
  7. 2024Confidence v.s. Critique: A Decomposition of Self-Correction Capability for LLMs
  8. 2025Self-Evolving Critique Abilities in Large Language Models