Self-Discover: Large Language Models Self-Compose Reasoning Structures
- URL: http://arxiv.org/abs/2402.03620v1
- Date: Tue, 6 Feb 2024 01:13:53 GMT
- Title: Self-Discover: Large Language Models Self-Compose Reasoning Structures
- Authors: Pei Zhou, Jay Pujara, Xiang Ren, Xinyun Chen, Heng-Tze Cheng, Quoc V.
Le, Ed H. Chi, Denny Zhou, Swaroop Mishra, Huaixiu Steven Zheng
- Abstract summary: We introduce SELF-DISCOVER, a framework for self-discovering task-intrinsic reasoning structures.
SELF-DISCOVER substantially improves GPT-4 and PaLM 2's performance on challenging reasoning benchmarks.
We show that the self-discovered reasoning structures are universally applicable across model families.
- Score: 136.48389510481758
- License: http://creativecommons.org/licenses/by/4.0/
- 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.
Related papers
- LLMs Are Not Intelligent Thinkers: Introducing Mathematical Topic Tree Benchmark for Comprehensive Evaluation of LLMs [8.89259409245068]
Large language models (LLMs) demonstrate impressive capabilities in mathematical reasoning.
We present the Mathematical Topics Tree (MaTT) benchmark, a challenging and structured benchmark that offers 1,958 questions.
We find that the most advanced model, GPT-4, achieved a mere 54% accuracy in a multiple-choice scenario.
arXiv Detail & Related papers (2024-06-07T18:21:26Z) - MindStar: Enhancing Math Reasoning in Pre-trained LLMs at Inference Time [51.5039731721706]
MindStar is a purely inference-based searching method for large language models.
It formulates reasoning tasks as searching problems and proposes two search ideas to identify the optimal reasoning paths.
It significantly enhances the reasoning abilities of open-source models, such as Llama-2-13B and Mistral-7B, and achieves comparable performance to GPT-3.5 and Grok-1.
arXiv Detail & Related papers (2024-05-25T15:07:33Z) - LLM Reasoners: New Evaluation, Library, and Analysis of Step-by-Step Reasoning with Large Language Models [25.537725151112387]
We introduce AutoRace for fully automated reasoning chain evaluation.
We also develop LLM Reasoners, a library for standardized modular implementation of existing and new reasoning algorithms.
arXiv Detail & Related papers (2024-04-08T06:35:09Z) - Direct Evaluation of Chain-of-Thought in Multi-hop Reasoning with Knowledge Graphs [52.42505579545893]
Large language models (LLMs) demonstrate strong reasoning abilities when prompted to generate chain-of-thought explanations alongside answers.
We propose a novel discriminative and generative CoT evaluation paradigm to assess LLMs' knowledge of reasoning and the accuracy of the generated CoT.
arXiv Detail & Related papers (2024-02-17T05:22:56Z) - SEER: Facilitating Structured Reasoning and Explanation via Reinforcement Learning [29.514755268807868]
We propose SEER, a novel method that maximizes a structure-based return to facilitate structured reasoning and explanation.
Our proposed structure-based return precisely describes the hierarchical and branching structure inherent in structured reasoning.
Our experiments show that SEER significantly outperforms state-of-the-art methods.
arXiv Detail & Related papers (2024-01-24T06:10:51Z) - A Principled Framework for Knowledge-enhanced Large Language Model [58.1536118111993]
Large Language Models (LLMs) are versatile, yet they often falter in tasks requiring deep and reliable reasoning.
This paper introduces a rigorously designed framework for creating LLMs that effectively anchor knowledge and employ a closed-loop reasoning process.
arXiv Detail & Related papers (2023-11-18T18:10:02Z) - Improving Open Information Extraction with Large Language Models: A
Study on Demonstration Uncertainty [52.72790059506241]
Open Information Extraction (OIE) task aims at extracting structured facts from unstructured text.
Despite the potential of large language models (LLMs) like ChatGPT as a general task solver, they lag behind state-of-the-art (supervised) methods in OIE tasks.
arXiv Detail & Related papers (2023-09-07T01:35:24Z) - Metacognitive Prompting Improves Understanding in Large Language Models [12.112914393948415]
We introduce Metacognitive Prompting (MP), a strategy inspired by human introspective reasoning processes.
We conduct experiments on four prevalent Large Language Models (LLMs) across ten natural language understanding (NLU) datasets.
MP consistently outperforms existing prompting methods in both general and domain-specific NLU tasks.
arXiv Detail & Related papers (2023-08-10T05:10:17Z) - Exploring Self-supervised Logic-enhanced Training for Large Language Models [59.227222647741094]
In this paper, we make the first attempt to investigate the feasibility of incorporating logical knowledge through self-supervised post-training.
We devise an auto-regressive objective variant of MERIt and integrate it with two LLM series, i.e., FLAN-T5 and LLaMA, with parameter size ranging from 3 billion to 13 billion.
The results on two challenging logical reasoning benchmarks demonstrate the effectiveness of LogicLLM.
arXiv Detail & Related papers (2023-05-23T06:13:10Z) - Selection-Inference: Exploiting Large Language Models for Interpretable
Logical Reasoning [14.663216851932646]
We show that language models tend to perform fairly well at single step inference tasks, but struggle to chain together multiple reasoning steps to solve more complex problems.
We propose a Selection-Inference (SI) framework that exploits pre-trained LLMs as general processing modules.
We show that a 7B parameter LLM used within the SI framework in a 5-shot generalisation setting, with no fine-tuning, yields a performance improvement of over 100%.
arXiv Detail & Related papers (2022-05-19T17:25:28Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.