mlsanityData to 14 Oct 2026

MoT: Memory-of-Thought Enables ChatGPT to Self-Improve

arXiv 2305.05181 ↗First posted 9 May 2023cs.CL cs.AIcited by 38

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

also in 1 more
Abstract (arXiv)

Large Language Models (LLMs) have shown impressive abilities in various tasks. However, fundamentally improving them depends on high-quality datasets or computationally expensive fine-tuning. On the contrary, humans can easily improve themselves by self-thinking and memory, without external resources. In this paper, we propose a framework, MoT, to let the LLM self-improve through Memory-of-Thought, without annotated datasets and parameter updates. Specifically, MoT is divided into two stages: 1. before the test stage, the LLM pre-thinks on the unlabeled dataset and saves the high-confidence thoughts as external memory; 2. During the test stage, given a test question, the LLM recalls relevant memory to help itself reason and answer it. Experimental results show that MoT can help ChatGPT significantly improve its abilities in arithmetic reasoning, commonsense reasoning, factual reasoning, and natural language inference. Further analyses show that each component contributes critically to the improvements and MoT can lead to consistent improvements across various CoT methods and LLMs.

Key facts. Cited by 38 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 48 of its 80 references that are in the map. First posted on arXiv 9 May 2023.

Builds on

48 of 80 references.

Closest ancestors · 3

Same direction, newest first

  1. 2022Large Language Models Can Self-Improve20 Oct
  2. 2022Language Models (Mostly) Know What They Know11 Jul
  3. 2022STaR: Bootstrapping Reasoning With Reasoning28 Mar

Other references · 45

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  4. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  5. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,326
  6. 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
  7. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
  8. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,512
  9. 2022Scaling Instruction-Finetuned Language Modelscited by 3,024
  10. 2022Training Compute-Optimal Large Language Modelscited by 3,019
  11. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
  12. 2023A Survey of Large Language Modelscited by 2,576
33 more
  1. 2022Emergent Abilities of Large Language Modelscited by 2,512
  2. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,998
  3. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,407
  4. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,075
  5. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 810
  6. 2022A Survey on In-context Learningcited by 721
  7. 2022Automatic Chain of Thought Prompting in Large Language Modelscited by 688
  8. 2022Towards Reasoning in Large Language Models: A Surveycited by 596
  9. 2023Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Modelscited by 560
  10. 2020Pre-trained Models for Natural Language Processing: A Surveycited by 505
  11. 2023Augmented Language Models: a Surveycited by 403
  12. 2022Unnatural Instructions: Tuning Language Models with (Almost) No Human Laborcited by 397
  13. 2022One Embedder, Any Task: Instruction-Finetuned Text Embeddingscited by 333
  14. 2022Large Language Models Are Reasoning Teacherscited by 329
  15. 2022Large Language Models are Better Reasoners with Self-Verificationcited by 307
  16. 2022UL2: Unifying Language Learning Paradigmscited by 295
  17. 2022Teaching Small Language Models to Reasoncited by 284
  18. 2023Specializing Smaller Language Models towards Multi-Step Reasoningcited by 251
  19. 2022Selective Annotation Makes Language Models Better Few-Shot Learnerscited by 243
  20. 2022In-context Examples Selection for Machine Translationcited by 172
  21. 2023Progressive-Hint Prompting Improves Reasoning in Large Language Modelscited by 147
  22. 2023Compositional Exemplars for In-context Learningcited by 144
  23. 2023Unified Demonstration Retriever for In-Context Learningcited by 142
  24. 2022Diverse Demonstrations Improve In-context Compositional Generalizationcited by 139
  25. 2022Rationale-Augmented Ensembles in Language Modelscited by 111
  26. 2023Finding Support Examples for In-Context Learningcited by 104
  27. 2022Z-ICL: Zero-Shot In-Context Learning with Pseudo-Demonstrationscited by 70
  28. 2023Synthetic Prompting: Generating Chain-of-Thought Demonstrations for Large Language Modelscited by 61
  29. 2022Self-Guided Noise-Free Data Generation for Efficient Zero-Shot Learningcited by 50
  30. 2022In-Context Learning for Few-Shot Dialogue State Trackingcited by 47
  31. 2022Self-Prompting Large Language Models for Zero-Shot Open-Domain QAcited by 39
  32. 2023Language Models are Few-shot Learners for Prognostic Predictioncited by 37
  33. 2022XRICL: Cross-lingual Retrieval-Augmented In-Context Learning for Cross-lingual Text-to-SQL Semantic Parsingcited by 33

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 2; 2024: 14; 2025: 14; 2026: 8.

38 papers.

Sort:

By area:

  1. 2024From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judgecited by 494
  2. 2024A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Modelscited by 452
  3. 2024LLM Multi-Agent Systems: Challenges and Open Problemscited by 120
  4. 2023Can Knowledge Graphs Reduce Hallucinations in LLMs? : A Surveycited by 92
  5. 2024A Survey on Self-Evolution of Large Language Modelscited by 91
  6. 2023Exchange-of-Thought: Enhancing Large Language Model Capabilities through Cross-Model Communicationcited by 82
  7. 2024Plan-on-Graph: Self-Correcting Adaptive Planning of Large Language Model on Knowledge Graphscited by 78
  8. 2024StreamBench: Towards Benchmarking Continuous Improvement of Language Agentscited by 47
  9. 2025Large Language Models for Planning: A Comprehensive and Systematic Surveycited by 40
  10. 2024What Factors Affect Multi-Modal In-Context Learning? An In-Depth Explorationcited by 31
  11. 2026On Safety Risks in Experience-Driven Self-Evolving Agentscited by 14
  12. 2024Unearthing Large Scale Domain-Specific Knowledge from Public Corporacited by 12
26 more
  1. 2024Importance Weighting Can Help Large Language Models Self-Improvecited by 12
  2. 2024Self-Evolving GPT: A Lifelong Autonomous Experiential Learnercited by 9
  3. 2025HiMA-Ecom: Enabling Joint Training of Hierarchical Multi-Agent E-commerce Assistantscited by 9
  4. 2025Scaling Laws for Many-Shot In-Context Learning with Self-Generated Annotationscited by 7
  5. 2025Multi-Agent LLM Judge: automatic personalized LLM judge design for evaluating natural language generation applicationscited by 7
  6. 2025Knowledge Augmented Complex Problem Solving with Large Language Models: A Surveycited by 6
  7. 2025Exploring Consciousness in LLMs: A Systematic Survey of Theories, Implementations, and Frontier Riskscited by 5
  8. 2025Formalizing the Safety, Security, and Functional Properties of Agentic AI Systemscited by 5
  9. 2026Large Language Model Agents Are Not Always Faithful Self-Evolverscited by 5
  10. 2024ONSEP: A Novel Online Neural-Symbolic Framework for Event Prediction Based on Large Language Modelcited by 4
  11. 2024Zero-to-Strong Generalization: Eliciting Strong Capabilities of Large Language Models Iteratively without Gold Labelscited by 4
  12. 2024Deploying Foundation Model Powered Agent Services: A Surveycited by 4
  13. 2025CDW-CoT: Clustered Distance-Weighted Chain-of-Thoughts Reasoningcited by 3
  14. 2026CoG: Controllable Graph Reasoning via Relational Blueprints and Failure-Aware Refinement over Knowledge Graphscited by 3
  15. 2024PICLe: Pseudo-Annotations for In-Context Learning in Low-Resource Named Entity Detectioncited by 2
  16. 2025Fine-Grained Emotion Recognition via In-Context Learningcited by 2
  17. 2025Making Large Language Models Better Reasoners with Orchestrated Streaming Experiencescited by 1
  18. 2025Can Large Language Models Integrate Spatial Data? Empirical Insights into Reasoning Strengths and Computational Weaknessescited by 1
  19. 2025Guided Navigation in Knowledge-Dense Environments: Structured Semantic Exploration with Guidance Graphscited by 1
  20. 2025How2: How to learn from procedural How-to questionscited by 1
  21. 2026MetaEvo: A Meta-Optimization Framework for Experience-Driven Agent Evolutioncited by 1
  22. 2026Metacognition in LLMs: Foundations, Progress, and Opportunitiescited by 1
  23. 2025The Impact of Role Design in In-Context Learning for Large Language Modelscited by 0
  24. 2026Training-Free Test-Time Contrastive Learning for Large Language Modelscited by 0
  25. 2026LLM-X: A Scalable Negotiation-Oriented Exchange for Communication Among Personal LLM Agentscited by 0
  26. 2026Retrieval-Based In-Context Learning: A Domain Adaptation Frameworkcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing
  2. 2024Large Language Models Can Self-Improve in Long-context Reasoning
  3. 2025Beyond Chains of Thought: Benchmarking Latent-Space Reasoning Abilities in Large Language Models
  4. 2023Think-in-Memory: Recalling and Post-thinking Enable LLMs with Long-Term Memory
  5. 2025Double-Checker: Enhancing Reasoning of Slow-Thinking LLMs via Self-Critical Fine-Tuning
  6. 2025Advancing Reasoning in Large Language Models: Promising Methods and Approaches
  7. 2024Self-Explore: Enhancing Mathematical Reasoning in Language Models with Fine-grained Rewards
  8. 2026State of Thought Enables Endogenous Reasoning