mlsanityData to 7 Oct 2026

STaR: Bootstrapping Reasoning With Reasoning

arXiv 2203.14465 ↗First posted 28 Mar 2022cs.LG cs.AI cs.CLcited by 895

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

also in 1 more
Abstract (arXiv)

Generating step-by-step "chain-of-thought" rationales improves language model performance on complex reasoning tasks like mathematics or commonsense question-answering. However, inducing language model rationale generation currently requires either constructing massive rationale datasets or sacrificing accuracy by using only few-shot inference. We propose a technique to iteratively leverage a small number of rationale examples and a large dataset without rationales, to bootstrap the ability to perform successively more complex reasoning. This technique, the "Self-Taught Reasoner" (STaR), relies on a simple loop: generate rationales to answer many questions, prompted with a few rationale examples; if the generated answers are wrong, try again to generate a rationale given the correct answer; fine-tune on all the rationales that ultimately yielded correct answers; repeat. We show that STaR significantly improves performance on multiple datasets compared to a model fine-tuned to directly predict final answers, and performs comparably to fine-tuning a 30× larger state-of-the-art language model on CommensenseQA. Thus, STaR lets a model improve itself by learning from its own generated reasoning.

Key facts. Cited by 895 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 34 of its 57 references that are in the map. First posted on arXiv 28 Mar 2022.

Builds on

34 of 57 references.

References · 34

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2017Proximal Policy Optimization Algorithmscited by 14,441
  4. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  5. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  6. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  7. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  8. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,747
  9. 2020The Pile: An 800GB Dataset of Diverse Text for Language Modelingcited by 2,346
  10. 2018CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledgecited by 1,793
  11. 2022Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?cited by 1,413
  12. 2022LaMDA: Language Models for Dialog Applicationscited by 1,218
22 more
  1. 2017Explanation in Artificial Intelligence: Insights from the Social Sciencescited by 1,201
  2. 2016ConceptNet 5.5: An Open Multilingual Graph of General Knowledgecited by 1,065
  3. 2022Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agentscited by 1,029
  4. 2021An Explanation of In-context Learning as Implicit Bayesian Inferencecited by 915
  5. 2021Show Your Work: Scratchpads for Intermediate Computation with Language Modelscited by 871
  6. 2020Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?cited by 590
  7. 2018e-SNLI: Natural Language Inference with Natural Language Explanationscited by 468
  8. 2020Generative Language Modeling for Automated Theorem Provingcited by 379
  9. 2019Explain Yourself! Leveraging Language Models for Commonsense Reasoningcited by 367
  10. 2017Thinking Fast and Slow with Deep Learning and Tree Searchcited by 245
  11. 2020Unsupervised Commonsense Question Answering with Self-Talkcited by 161
  12. 2022Formal Mathematics Statement Curriculum Learningcited by 141
  13. 2014From Captions to Visual Concepts and Backcited by 109
  14. 2021Few-Shot Self-Rationalization with Natural Language Promptscited by 83
  15. 2017The Promise and Peril of Human Evaluation for Model Interpretabilitycited by 52
  16. 2022Improving Intrinsic Exploration with Language Abstractionscited by 49
  17. 2017From Language to Programs: Bridging Reinforcement Learning and Maximum Marginal Likelihoodcited by 41
  18. 2021Human Parity on CommonsenseQA: Augmenting Self-Attention with External Attentioncited by 40
  19. 2020Towards Interpretable Natural Language Understanding with Explanations as Latent Variablescited by 30
  20. 2021Generate Natural Language Explanations for Recommendationcited by 22
  21. 2021Iterated learning for emergent systematicity in VQAcited by 21
  22. 2017Weakly-supervised Semantic Parsing with Abstract Examplescited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 48; 2023: 108; 2024: 134; 2025: 286; 2026: 319.

895 papers.

Sort:

By area:

  1. 2022Large Language Models are Zero-Shot Reasonerscited by 4,323
  2. 2023Let's Verify Step by Stepcited by 3,598
  3. 2022Scaling Instruction-Finetuned Language Modelscited by 2,997
  4. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
  5. 2023Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Promptingcited by 1,081
  6. 2022Large Language Models Are Human-Level Prompt Engineerscited by 1,020
  7. 2025Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?cited by 948
  8. 2022Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questionscited by 911
  9. 2023Graph of Thoughts: Solving Elaborate Problems with Large Language Modelscited by 873
  10. 2022Solving math word problems with process- and outcome-based feedbackcited by 871
  11. 2023DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelinescited by 803
  12. 2023Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizescited by 710
38 more
  1. 2022Automatic Chain of Thought Prompting in Large Language Modelscited by 684
  2. 2023Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Modelscited by 659
  3. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 655
  4. 2023WizardMath: Empowering Mathematical Reasoning for Large Language Models via Reinforced Evol-Instructcited by 618
  5. 2023ChemCrow: Augmenting large-language models with chemistry toolscited by 609
  6. 2022Towards Reasoning in Large Language Models: A Surveycited by 593
  7. 2024Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMscited by 564
  8. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 554
  9. 2025Kimi K2: Open Agentic Intelligencecited by 523
  10. 2023Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolutioncited by 521
  11. 2023Is ChatGPT a General-Purpose Natural Language Processing Task Solver?cited by 486
  12. 2023Reinforced Self-Training (ReST) for Language Modelingcited by 486
  13. 2024Generative Verifiers: Reward Modeling as Next-Token Predictioncited by 465
  14. 2025CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Modelscited by 440
  15. 2023Cognitive Architectures for Language Agentscited by 422
  16. 2023Scaling Relationship on Learning Mathematical Reasoning with Large Language Modelscited by 417
  17. 2024Training Language Models to Self-Correct via Reinforcement Learningcited by 405
  18. 2025Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attentioncited by 391
  19. 2025Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRscited by 373
  20. 2023Language Models can Solve Computer Taskscited by 371
  21. 2024Automated Design of Agentic Systemscited by 360
  22. 2024ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Searchcited by 338
  23. 2025Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscationcited by 338
  24. 2022Selection-Inference: Exploiting Large Language Models for Interpretable Logical Reasoningcited by 331
  25. 2023Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Modelscited by 325
  26. 2025Absolute Zero: Reinforced Self-play Reasoning with Zero Datacited by 323
  27. 2025From System 1 to System 2: A Survey of Reasoning Large Language Modelscited by 319
  28. 2022Large Language Models Are Reasoning Teacherscited by 318
  29. 2022Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLPcited by 316
  30. 2025The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexitycited by 298
  31. 2023ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMscited by 297
  32. 2024Quiet-STaR: Language Models Can Teach Themselves to Think Before Speakingcited by 292
  33. 2023Alphazero-like Tree-Search can Guide Large Language Model Decoding and Trainingcited by 291
  34. 2026Reinforcement Learning via Self-Distillationcited by 285
  35. 2023ToRA: A Tool-Integrated Reasoning Agent for Mathematical Problem Solvingcited by 276
  36. 2024Rewarding Progress: Scaling Automated Process Verifiers for LLM Reasoningcited by 267
  37. 2024Personal LLM Agents: Insights and Survey about the Capability, Efficiency and Securitycited by 257
  38. 2024HuatuoGPT-o1, Towards Medical Complex Reasoning with LLMscited by 257

50 most cited of 895 All 895 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025HS-STaR: Hierarchical Sampling for Self-Taught Reasoners via Difficulty Estimation and Budget Reallocation
  2. 2024Self-Training Meets Consistency: Improving LLMs' Reasoning with Consistency-Driven Rationale Evaluation
  3. 2022Large Language Models Can Self-Improve
  4. 2024Small Language Models Need Strong Verifiers to Self-Correct Reasoning
  5. 2025The Self-Improvement Paradox: Can Language Models Bootstrap Reasoning Capabilities without External Scaffolding?
  6. 2025Socratic-Zero : Bootstrapping Reasoning via Data-Free Agent Co-evolution
  7. 2023Boosting the Power of Small Multimodal Reasoning Models to Match Larger Models with Self-Consistency Training
  8. 2025Towards Efficient CoT Distillation: Self-Guided Rationale Selector for Better Performance with Fewer Rationales