{"work":{"id":"3339dfe0-b229-41ec-a6e0-8f601080eb97","openalex_id":"https://openalex.org/W4226369848","doi":"10.48550/arxiv.2203.14465","arxiv_id":"2203.14465","raw_key":null,"title":"STaR: Bootstrapping Reasoning With Reasoning","authors":null,"authors_text":"Eric Zelikman, Yuhuai Wu, Jesse Mu, and Noah D","year":2022,"venue":"cs.LG","abstract":"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$\\times$ larger state-of-the-art language model on CommensenseQA. Thus, STaR lets a model improve itself by learning from its own generated reasoning.","external_url":"https://arxiv.org/abs/2203.14465","cited_by_count":116,"metadata_source":"pith","metadata_fetched_at":"2026-08-05T02:28:24.338817+00:00","pith_arxiv_id":"2203.14465","created_at":"2026-05-09T23:54:45.373497+00:00","updated_at":"2026-08-05T02:28:24.338817+00:00","title_quality_ok":true,"display_title":"Wu, andN.D.Goodman","render_title":"Wu, andN.D.Goodman"},"hub":{"state":{"work_id":"3339dfe0-b229-41ec-a6e0-8f601080eb97","tier":"hub","tier_reason":"10+ Pith inbound or 1,000+ external citations","pith_inbound_count":39,"external_cited_by_count":116,"distinct_field_count":5,"first_pith_cited_at":"2022-01-28T02:33:07+00:00","last_pith_cited_at":"2026-07-08T17:19:50+00:00","author_build_status":"not_needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"not_needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-08-22T05:29:28.325178+00:00","tier_text":"hub"},"tier":"hub","role_counts":[{"context_role":"background","n":3}],"polarity_counts":[{"context_polarity":"background","n":3}],"runs":{},"summary":{},"graph":{},"authors":[]}}