mlsanityData to 7 Oct 2026

Internal Consistency and Self-Feedback in Large Language Models: A Survey

arXiv 2407.14507 ↗First posted 19 Jul 2024cs.CLcited by 49

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

Abstract (arXiv)

Large language models (LLMs) often exhibit deficient reasoning or generate hallucinations. To address these, studies prefixed with "Self-" such as Self-Consistency, Self-Improve, and Self-Refine have been initiated. They share a commonality: involving LLMs evaluating and updating themselves. Nonetheless, these efforts lack a unified perspective on summarization, as existing surveys predominantly focus on categorization. In this paper, we use a unified perspective of internal consistency, offering explanations for reasoning deficiencies and hallucinations. Internal consistency refers to the consistency in expressions among LLMs' latent, decoding, or response layers based on sampling methodologies. Then, we introduce an effective theoretical framework capable of mining internal consistency, named Self-Feedback. This framework consists of two modules: Self-Evaluation and Self-Update. The former captures internal consistency signals, while the latter leverages the signals to enhance either the model's response or the model itself. This framework has been employed in numerous studies. We systematically classify these studies by tasks and lines of work; summarize relevant evaluation methods and benchmarks; and delve into the concern, "Does Self-Feedback Really Work?" We also propose several critical viewpoints, including the "Hourglass Evolution of Internal Consistency", "Consistency Is (Almost) Correctness" hypothesis, and "The Paradox of Latent and Explicit Reasoning". The relevant resources are open-sourced at https://github.com/IAAR-Shanghai/ICSFSurvey.

Key facts. Cited by 49 papers in the map (data through 7 Oct 2026). Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning. Builds on 142 of its 161 references that are in the map. First posted on arXiv 19 Jul 2024.

Builds on

142 of 161 references.

Closest ancestors · 13

Same direction, newest first

  1. 2024Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing18 Apr
  2. 2024CriticBench: Benchmarking LLMs for Critique-Correct Reasoning22 Feb
  3. 2024Soft Self-Consistency Improves Language Model Agents20 Feb
  4. 2024Pride and Prejudice: LLM Amplifies Self-Bias in Self-Refinement18 Feb
  5. 2023Universal Self-Consistency for Large Language Model Generation29 Nov
  6. 2023MAF: Multi-Aspect Feedback for Improving Reasoning in Large Language Models19 Oct
  7. 2023Large Language Models Cannot Self-Correct Reasoning Yet3 Oct
  8. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Models17 May
  9. 2023Self-Evaluation Guided Beam Search for Reasoning1 May
  10. 2023REFINER: Reasoning Feedback on Intermediate Representations4 Apr
  11. 2023Self-Refine: Iterative Refinement with Self-Feedback30 Mar
  12. 2022Large Language Models Can Self-Improve20 Oct
1 more
  1. 2022Language Models (Mostly) Know What They Know11 Jul

Other references · 129

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2017Proximal Policy Optimization Algorithmscited by 14,441
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  4. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  5. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  6. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  7. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  8. 2020Scaling Laws for Neural Language Modelscited by 6,411
  9. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  10. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  11. 2018Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challengecited by 4,710
  12. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
38 more
  1. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  2. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  3. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
  4. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,919
  5. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  6. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,674
  7. 2023A Survey of Large Language Modelscited by 2,537
  8. 2023A Survey on Large Language Model based Autonomous Agentscited by 2,203
  9. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,185
  10. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,767
  11. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
  12. 2023Inference-Time Intervention: Eliciting Truthful Answers from a Language Modelcited by 1,259
  13. 2023A Survey on Evaluation of Large Language Modelscited by 1,227
  14. 2023Teaching Large Language Models to Self-Debugcited by 963
  15. 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Modelscited by 873
  16. 2023On-Policy Distillation of Language Models: Learning from Self-Generated Mistakescited by 840
  17. 2023BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Datasetcited by 828
  18. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 787
  19. 2023Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formattingcited by 778
  20. 2022Discovering Latent Knowledge in Language Models Without Supervisioncited by 733
  21. 2021BARTScore: Evaluating Generated Text as Text Generationcited by 719
  22. 2023OpenAssistant Conversations -- Democratizing Large Language Model Alignmentcited by 712
  23. 2020Aligning AI With Shared Human Valuescited by 706
  24. 2023Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Modelscited by 695
  25. 2023C-Eval: A Multi-Level Multi-Discipline Chinese Evaluation Suite for Foundation Modelscited by 694
  26. 2022Contrastive Decoding: Open-ended Text Generation as Optimizationcited by 598
  27. 2024Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Modelscited by 552
  28. 2024Scaling and evaluating sparse autoencoderscited by 530
  29. 2023Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolutioncited by 521
  30. 2022Self-critiquing models for assisting human evaluatorscited by 372
  31. 2018WiC: the Word-in-Context Dataset for Evaluating Context-Sensitive Meaning Representationscited by 370
  32. 2023DoLa: Decoding by Contrasting Layers Improves Factuality in Large Language Modelscited by 367
  33. 2021Measuring and Improving Consistency in Pretrained Language Modelscited by 361
  34. 2023GPTScore: Evaluate as You Desirecited by 357
  35. 2023Trusting Your Evidence: Hallucinate Less with Context-aware Decodingcited by 319
  36. 2023Principle-Driven Self-Alignment of Language Models from Scratch with Minimal Human Supervisioncited by 318
  37. 2024INSIDE: LLMs' Internal States Retain the Power of Hallucination Detectioncited by 317
  38. 2023How Language Model Hallucinations Can Snowballcited by 310

50 most cited of 129 All 129 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 13; 2025: 27; 2026: 9.

49 papers.

Sort:

By area:

  1. 2025Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Modelscited by 381
  2. 2024LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinationscited by 230
  3. 2025Scalable Best-of-N Selection for Large Language Models via Self-Certaintycited by 182
  4. 2024Attention Heads of Large Language Models: A Surveycited by 65
  5. 2024Latent Space Chain-of-Embedding Enables Output-free LLM Self-Evaluationcited by 56
  6. 2025xVerify: Efficient Answer Verifier for Reasoning Model Evaluationscited by 53
  7. 2024Self-Consistency Preference Optimizationcited by 50
  8. 2024Unveiling Visual Perception in Language Models: An Attention Head Analysis Approachcited by 34
  9. 2025Entropy-based Exploration Conduction for Multi-step Reasoningcited by 28
  10. 2024Surveying the MLLM Landscape: A Meta-Review of Current Surveyscited by 20
  11. 2025MoC: Mixtures of Text Chunking Learners for Retrieval-Augmented Generation Systemcited by 19
  12. 2025Large language models for artificial general intelligence (AGI): A survey of foundational principles and approachescited by 18
37 more
  1. 2024Meta-Chunking: Learning Text Segmentation and Semantic Completion via Logical Perceptioncited by 17
  2. 2025SAND: Boosting LLM Agents with Self-Taught Action Deliberationcited by 14
  3. 2025Alignment and Safety in Large Language Models: Safety Mechanisms, Training Paradigms, and Emerging Challengescited by 14
  4. 2024Benchmarking the Capabilities of Large Language Models in Transportation System Engineering: Accuracy, Consistency, and Reasoning Behaviorscited by 13
  5. 2025ToMAP: Training Opponent-Aware LLM Persuaders with Theory of Mindcited by 12
  6. 2025Learning from Diverse Reasoning Paths with Routing and Collaborationcited by 12
  7. 2025Co-Sight: Enhancing LLM-Based Agents via Conflict-Aware Meta-Verification and Trustworthy Reasoning with Structured Factscited by 12
  8. 2024Seeking Consistent Flat Minima for Better Domain Generalization via Refining Loss Landscapescited by 10
  9. 2025Adaptive Termination for Multi-round Parallel Reasoning: An Universal Semantic Entropy-Guided Frameworkcited by 10
  10. 2025Dynamic Attention-Guided Context Decoding for Mitigating Context Faithfulness Hallucinations in Large Language Modelscited by 9
  11. 2025BioAgents: Democratizing Bioinformatics Analysis with Multi-Agent Systemscited by 8
  12. 2024TurtleBench: Evaluating Top Language Models via Real-World Yes/No Puzzlescited by 7
  13. 2024Blind Spot Navigation in Large Language Model Reasoning with Thought Space Explorercited by 7
  14. 2025GRAPHMOE: Amplifying Cognitive Depth of Mixture-of-Experts Network via Introducing Self-Rethinking Mechanismcited by 7
  15. 2026REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Tracescited by 7
  16. 2025Loki's Dance of Illusions: A Comprehensive Survey of Hallucination in Large Language Modelscited by 6
  17. 2025SHARP: Synthesizing High-quality Aligned Reasoning Problems for Large Reasoning Models Reinforcement Learningcited by 5
  18. 2025Survey of GenAI for Automotive Software Development: From Requirements to Executable Codecited by 5
  19. 2024KBAlign: Efficient Self Adaptation on Specific Knowledge Basescited by 3
  20. 2025Temperature and Persona Shape LLM Agent Consensus With Minimal Accuracy Gains in Qualitative Codingcited by 3
  21. 2026PRISM: Probing Reasoning, Instruction, and Source Memory in LLM Hallucinationscited by 3
  22. 2025DSVD: Dynamic Self-Verify Decoding for Faithful Generation in Large Language Modelscited by 2
  23. 2025Latent Debate: A Surrogate Framework for Interpreting LLM Thinkingcited by 2
  24. 2026Large Language Models Decide Early and Explain Latercited by 2
  25. 2025MoM: Mixtures of Scenario-Aware Document Memories for Retrieval-Augmented Generation Systemscited by 1
  26. 2025Simulated Self-Assessment in Large Language Models: A Psychometric Approach to AI Self-Efficacycited by 1
  27. 2025Multi-chain Graph Refinement and Selection for Reliable Reasoning in Large Language Modelscited by 1
  28. 2024Evolution of Thought: Diverse and High-Quality Reasoning via Multi-Objective Optimizationcited by 0
  29. 2025Why Stop at One Error? Benchmarking LLMs as Data Science Code Debuggers for Multi-Hop and Multi-Bug Errorscited by 0
  30. 2025Explainability Through Systematicity: The Hard Systematicity Challenge for Artificial Intelligencecited by 0
  31. 2025Mastering Olympiad-Level Physics with Artificial Intelligencecited by 0
  32. 2026FVRuleLearner: Operator-Level Reasoning Tree (Op-Tree)-Based Rules Learning for Formal Verificationcited by 0
  33. 2026Reasoning Portability: Guiding Continual Learning for MLLMs in the RLVR Eracited by 0
  34. 2026Pseudocode-Guided Structured Reasoning for Automating Reliable Inference in Vision-Language Modelscited by 0
  35. 2026Customized Generative AI Agent for Transportation Engineering Practice: A Development and Continued Pre-training Guidelinecited by 0
  36. 2026Simulation-Aware In-Context Policy Improvement for LLM-Aided Analog Layout Refinementcited by 0
  37. 2026ChainPrune: Evaluating and Reducing Redundancy in Long Chain-of-Thought Reasoningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Self-Contrast: Better Reflection Through Inconsistent Solving Perspectives
  2. 2025Self-Correcting Large Language Models: Generation vs. Multiple Choice
  3. 2023Self-Consistency of Large Language Models under Ambiguity
  4. 2025Self-Consistency Is Losing Its Edge: Diminishing Returns and Rising Costs in Modern LLMs
  5. 2024Mind the Gap: Examining the Self-Improvement Capabilities of Large Language Models
  6. 2024Semantic Self-Consistency: Enhancing Language Model Reasoning via Semantic Weighting
  7. 2025Language Models can perform Single-Utterance Self-Correction of Perturbed Reasoning
  8. 2023Two Failures of Self-Consistency in the Multi-Step Reasoning of LLMs