2022 STaRBootstrappingReasoningwithR
- (Zelikman et al., 2022) ⇒ Eric Zelikman, Yuhuai Wu, Jesse Mu, and Noah Goodman. (2022). “STaR: Bootstrapping Reasoning with Reasoning.” In: Advances in Neural Information Processing Systems, 35.
Subject Headings: Self-Taught Reasoner (STaR) Algorithm.
Notes
Cited By
Quotes
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× larger state-of-the-art language model on CommensenseQA. Thus, STaR lets a model improve itself by learning from its own generated reasoning.
References
;
| Author | volume | Date Value | title | type | journal | titleUrl | doi | note | year | |
|---|---|---|---|---|---|---|---|---|---|---|
| 2022 STaRBootstrappingReasoningwithR | Noah D. Goodman Yuhuai Wu Eric Zelikman Jesse Mu | Star: Bootstrapping Reasoning with Reasoning | 2022 |