mlsanityData to 7 Oct 2026

The CoT Collection: Improving Zero-shot and Few-shot Learning of Language Models via Chain-of-Thought Fine-Tuning

arXiv 2305.14045 ↗First posted 23 May 2023cs.CL cs.AI cs.LGcited by 106

On the map: LLM reasoning → Prompting and chain-of-thought reasoning

Abstract (arXiv)

Language models (LMs) with less than 100B parameters are known to perform poorly on chain-of-thought (CoT) reasoning in contrast to large LMs when solving unseen tasks. In this work, we aim to equip smaller LMs with the step-by-step reasoning capability by instruction tuning with CoT rationales. In order to achieve this goal, we first introduce a new instruction-tuning dataset called the CoT Collection, which augments the existing Flan Collection (including only 9 CoT tasks) with additional 1.84 million rationales across 1,060 tasks. We show that CoT fine-tuning Flan-T5 (3B & 11B) with CoT Collection enables smaller LMs to have better CoT capabilities on unseen tasks. On the BIG-Bench-Hard (BBH) benchmark, we report an average improvement of +4.34% (Flan-T5 3B) and +2.60% (Flan-T5 11B), in terms of zero-shot task accuracy. Furthermore, we show that instruction tuning with CoT Collection allows LMs to possess stronger few-shot learning capabilities on 4 domain-specific tasks, resulting in an improvement of +2.24% (Flan-T5 3B) and +2.37% (Flan-T5 11B), even outperforming ChatGPT utilizing demonstrations until the max length by a +13.98% margin. Our code, the CoT Collection data, and model checkpoints are publicly available.

Key facts. Cited by 106 papers in the map (data through 7 Oct 2026). Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning. Builds on 65 of its 98 references that are in the map. First posted on arXiv 23 May 2023.

Builds on

65 of 98 references.

Closest ancestors · 8

Same direction, newest first

  1. 2023CoTEVer: Chain of Thought Prompting Annotation Toolkit for Explanation Verification7 Mar
  2. 2022Large Language Models Are Reasoning Teachers20 Dec
  3. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them17 Oct
  4. 2022Automatic Chain of Thought Prompting in Large Language Models7 Oct
  5. 2022Large Language Models are Zero-Shot Reasoners24 May
  6. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Models21 May
  7. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Models21 Mar
  8. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Models28 Jan

Other references · 57

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2022Training language models to follow instructions with human feedbackcited by 15,712
  3. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  4. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  5. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  6. 2020Scaling Laws for Neural Language Modelscited by 6,411
  7. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  8. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  9. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  10. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  11. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,155
  12. 2022Scaling Instruction-Finetuned Language Modelscited by 2,997
38 more
  1. 2019The Curious Case of Neural Text Degenerationcited by 2,881
  2. 2020mT5: A massively multilingual pre-trained text-to-text transformercited by 1,813
  3. 2021Multitask Prompted Training Enables Zero-Shot Task Generalizationcited by 1,462
  4. 2019PubMedQA: A Dataset for Biomedical Research Question Answeringcited by 1,271
  5. 2020Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inferencecited by 934
  6. 2022STaR: Bootstrapping Reasoning With Reasoningcited by 895
  7. 2022Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learningcited by 886
  8. 2022Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Taskscited by 858
  9. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 736
  10. 2023The Flan Collection: Designing Data and Methods for Effective Instruction Tuningcited by 734
  11. 2021GPT Understands, Toocited by 715
  12. 2021Cross-Task Generalization via Natural Language Crowdsourcing Instructionscited by 696
  13. 2022Large Language Models Can Self-Improvecited by 656
  14. 2019MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalismscited by 653
  15. 2021Datasets: A Community Library for Natural Language Processingcited by 526
  16. 2022Language Models are Multilingual Chain-of-Thought Reasonerscited by 518
  17. 2018e-SNLI: Natural Language Inference with Natural Language Explanationscited by 468
  18. 2021MetaICL: Learning to Learn In Contextcited by 450
  19. 2022Unnatural Instructions: Tuning Language Models with (Almost) No Human Laborcited by 395
  20. 2023Multi-step Jailbreaking Privacy Attacks on ChatGPTcited by 320
  21. 2022UL2: Unifying Language Learning Paradigmscited by 292
  22. 2021Symbolic Knowledge Distillation: from General Language Models to Commonsense Modelscited by 270
  23. 2023Specializing Smaller Language Models towards Multi-Step Reasoningcited by 249
  24. 2019QASC: A Dataset for Question Answering via Sentence Compositioncited by 248
  25. 2022Can language models learn from explanations in context?cited by 248
  26. 2022OPT-IML: Scaling Language Model Instruction Meta Learning through the Lens of Generalizationcited by 235
  27. 2022WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creationcited by 200
  28. 2021When Does Pretraining Help? Assessing Self-Supervised Learning for Law and the CaseHOLD Datasetcited by 199
  29. 2018Constituency Parsing with a Self-Attentive Encodercited by 196
  30. 2022ROSCOE: A Suite of Metrics for Scoring Step-by-Step Reasoningcited by 175
  31. 2021Towards Continual Knowledge Learning of Language Modelscited by 172
  32. 2020Recall and Learn: Fine-tuning Deep Pretrained Language Models with Less Forgettingcited by 169
  33. 2022SODA: Million-scale Dialogue Distillation with Social Commonsense Contextualizationcited by 166
  34. 2021ExT5: Towards Extreme Multi-Task Scaling for Transfer Learningcited by 161
  35. 2018Lessons from Natural Language Inference in the Clinical Domaincited by 149
  36. 2018Multilingual Constituency Parsing with Self-Attention and Pre-Trainingcited by 117
  37. 2020Graph-based, Self-Supervised Program Repair from Diagnostic Feedbackcited by 93
  38. 2021CREAK: A Dataset for Commonsense Reasoning over Entity Knowledgecited by 88

50 most cited of 57 All 57 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 10; 2024: 43; 2025: 42; 2026: 11.

106 papers.

Sort:

By area:

  1. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,185
  2. 2023A Comprehensive Overview of Large Language Modelscited by 639
  3. 2023Prometheus: Inducing Fine-grained Evaluation Capability in Language Modelscited by 558
  4. 2023FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Setscited by 198
  5. 2023Navigate through Enigmatic Labyrinth A Survey of Chain of Thought Reasoning: Advances, Frontiers and Futurecited by 184
  6. 2024Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuningcited by 174
  7. 2024Prometheus-Vision: Vision-Language Model as a Judge for Fine-Grained Evaluationcited by 123
  8. 2024On Memorization of Large Language Models in Logical Reasoningcited by 114
  9. 2023Can Knowledge Graphs Reduce Hallucinations in LLMs? : A Surveycited by 92
  10. 2024Towards Building Multilingual Language Model for Medicinecited by 91
  11. 2025Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoningcited by 87
  12. 2024Can LLMs Learn from Previous Mistakes? Investigating LLMs' Errors to Boost for Reasoningcited by 69
38 more
  1. 2024Training Large Language Models for Reasoning through Reverse Curriculum Reinforcement Learningcited by 63
  2. 2023Guiding Language Model Reasoning with Planning Tokenscited by 58
  3. 2025Grounded Chain-of-Thought for Multimodal Large Language Modelscited by 57
  4. 2025One Token to Fool LLM-as-a-Judgecited by 56
  5. 2023Prompt Engineering a Prompt Engineercited by 53
  6. 2024Usable XAI: 10 Strategies Towards Exploiting Explainability in the LLM Eracited by 45
  7. 2024MACM: Utilizing a Multi-Agent System for Condition Mining in Solving Complex Mathematical Problemscited by 41
  8. 2025Recent Advances in Large Langauge Model Benchmarks against Data Contamination: From Static to Dynamic Evaluationcited by 38
  9. 2024A Preliminary Study of o1 in Medicine: Are We Closer to an AI Doctor?cited by 33
  10. 2024Evaluating Language Models as Synthetic Data Generatorscited by 32
  11. 2024Self-Explore: Enhancing Mathematical Reasoning in Language Models with Fine-grained Rewardscited by 31
  12. 2023MUFFIN: Curating Multi-Faceted Instructions for Improving Instruction-Followingcited by 30
  13. 2024PCA-Bench: Evaluating Multimodal Large Language Models in Perception-Cognition-Action Chaincited by 29
  14. 2024From Sparse Dependence to Sparse Attention: Unveiling How Chain-of-Thought Enhances Transformer Sample Efficiencycited by 28
  15. 2024On Behalf of the Stakeholders: Trends in NLP Model Interpretability in the Era of LLMscited by 24
  16. 2025Robot-R1: Reinforcement Learning for Enhanced Embodied Reasoning in Roboticscited by 21
  17. 2024Fine-tuning Large Language Models with Sequential Instructionscited by 20
  18. 2024Can Knowledge Graphs Make Large Language Models More Trustworthy? An Empirical Study Over Open-ended Question Answeringcited by 20
  19. 2024Instruction Diversity Drives Generalization To Unseen Taskscited by 15
  20. 2024Optimizing Language Model's Reasoning Abilities with Weak Supervisioncited by 15
  21. 2024Cross-Data Knowledge Graph Construction for LLM-enabled Educational Question-Answering System: A Case Study at HCMUTcited by 14
  22. 2024What are the Essential Factors in Crafting Effective Long Context Multi-Hop Instruction Datasets? Insights and Best Practicescited by 14
  23. 2025Unmasking Deceptive Visuals: Benchmarking Multimodal Large Language Models on Misleading Chart Question Answeringcited by 14
  24. 2025BigCharts-R1: Enhanced Chart Reasoning with Visual Reinforcement Finetuningcited by 13
  25. 2023Meta-Reasoning: Semantics-Symbol Deconstruction for Large Language Modelscited by 11
  26. 2025Compositional Generalization from Learned Skills via CoT Training: A Theoretical and Structural Analysis for Reasoningcited by 10
  27. 2025Sherlock: Reliable and Efficient Agentic Workflow Executioncited by 10
  28. 2024Injecting Salesperson's Dialogue Strategies in Large Language Models with Chain-of-Thought Reasoningcited by 9
  29. 2024ChainLM: Empowering Large Language Models with Improved Chain-of-Thought Promptingcited by 8
  30. 2024TAIA: Large Language Models are Out-of-Distribution Data Learnerscited by 8
  31. 2024Chain-of-Though (CoT) prompting strategies for medical error detection and correctioncited by 7
  32. 2024How language models extrapolate outside the training data: A case study in Textualized Gridworldcited by 7
  33. 2024Prompting and Fine-tuning Large Language Models for Automated Code Review Comment Generationcited by 7
  34. 2025Select2Reason: Efficient Instruction-Tuning Data Selection for Long-CoT Reasoningcited by 7
  35. 2025Improving Rationality in the Reasoning Process of Language Models through Self-playing Gamecited by 7
  36. 2024Instruction Matters: A Simple yet Effective Task Selection for Optimized Instruction Tuning of Specific Taskscited by 6
  37. 2024Generative Prompt Internalizationcited by 6
  38. 2024Thinking with Knowledge Graphs: Enhancing LLM Reasoning Through Structured Datacited by 6

50 most cited of 106 All 106 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025CoT Vectors: Transferring and Probing the Reasoning Mechanisms of LLMs
  2. 2025Reasoning Transfer for an Extremely Low-Resource and Endangered Language: Bridging Languages Through Sample-Efficient Language Understanding
  3. 2025The Challenge of Teaching Reasoning to LLMs Without RL or Distillation
  4. 2025When Thinking Fails: The Pitfalls of Reasoning for Instruction-Following in LLMs
  5. 2024Fine-Tuning on Diverse Reasoning Chains Drives Within-Inference CoT Refinement in LLMs
  6. 2026S3-CoT: Self-Sampled Succinct Reasoning Enables Efficient Chain-of-Thought LLMs
  7. 2024On the Hardness of Faithful Chain-of-Thought Reasoning in Large Language Models
  8. 2026Zipping the Thought: When and How Compressed Reasoning Data Works in LLM Post-Training