SELF-DISCOVER: Large Language Models Self-Compose Reasoning Structures

Pei Zhou, Jay Pujara, Xiang Ren, Xinyun Chen, Heng-Tze Cheng, Quoc V Le, Ed H. Chi, Denny Zhou, Swaroop Mishra, Steven Zheng

Keywords: Large Language Models, Reasoning, Prompting, Self-Improve, Self-Discover

TL;DR: We introduce SELF-DISCOVER, an efficient and performant framework for models to self-discover a reasoning structure for any task from a seed set of general problem-solving skills.

Abstract: We introduce SELF-DISCOVER, a general framework for LLMs to self-discover the task-intrinsic reasoning structures to tackle complex reasoning problems that are challenging for typical prompting methods. Core to the framework is a self-discovery process where LLMs select multiple atomic reasoning modules such as critical thinking and step-by-step thinking, and compose them into an explicit reasoning structure for LLMs to follow during decoding. SELF-DISCOVER substantially improves GPT-4 and PaLM 2’s performance on challenging reasoning benchmarks such as BigBench-Hard, grounded agent reasoning, and MATH, by as much as 32% compared to Chain of Thought (CoT). Furthermore, SELF-DISCOVER outperforms inference-intensive methods such as CoT-Self-Consistency by more than 20%, while requiring 10-40x fewer inference compute. Finally, we show that the self-discovered reasoning structures are universally applicable across model families: from PaLM 2-L to GPT-4, and from GPT-4 to Llama2, and share commonalities with human reasoning patterns.

Primary Area: Natural language processing

Submission Number: 17915