Thought-Like-Pro：用自举式 Prolog 链式推理提升大语言模型推理能力
Thought-Like-Pro: Enhancing Reasoning of Large Language Models through Self-Bootstrapped Prolog-based Chain-of-Thought
Thought-Like-Pro 是一个结合符号 Prolog 逻辑引擎与大语言模型的模仿学习框架，通过让模型自主生成规则并借助 Prolog 推导推理轨迹，再转写为自然语言链式推理进行模仿学习。该框架以提示引导但自举的方式运行，显著提升了模型推理能力，并在分布外泛化任务上仅出现轻微性能下降。所用部分数据集已开源，论文被 IEEE TCDS 接收。
Authors:Xihe Qiu (1 and 2), Yongxin Deng (1 and 3), Xiaoyu Tan (2), Zhen Fang (3), Ling Chen (3) ((1) Shanghai University of Engineering Science, (2) National University of Singapore, (3) University of Technology Sydney)
Abstract:Large language models have demonstrated remarkable capabilities as general-purpose assistants, excelling in a wide range of reasoning tasks and supporting various aspects of daily web usage. This achievement represents a significant step toward achieving artificial general intelligence. Despite these advancements, the effectiveness of large language models often hinges on the specific prompting strategies employed, and there remains a lack of a robust framework to facilitate learning and generalization across diverse reasoning tasks. To address these challenges, we introduce a novel learning framework, Thought-Like-Pro. In this framework, we utilize imitation learning to imitate the Chain-of-Thought process which is verified and translated from reasoning trajectories generated by a symbolic Prolog logic engine. This framework proceeds in a prompt-guided but self-bootstrapped manner, that enables large language models to formulate rules and statements from given instructions and leverage the symbolic Prolog engine to derive results. Subsequently, large language models convert Prolog-derived successive reasoning trajectories into natural language chain-of-thought for imitation learning. The empirical findings indicate that our proposed approach greatly improves the reasoning capacity of large language models. By employing model averaging techniques, our method exhibits only a marginal decline in performance for distributional extrapolation tasks, showing robust generalization capabilities. We present a technical approach that integrates symbolic reasoning with language modeling, with the potential to support the development of large language models as cognitively inspired systems. The part of the dataset we used has been open-sourced.
| Comments: | 15 pages, including appendices. Accepted for publication in IEEE Transactions on Cognitive and Developmental Systems (TCDS) |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| MSC classes: | 68T50 (Primary), 68T27, 68T05 (Secondary) |
| ACM classes: | I.2.7; I.2.3; I.2.6 |
| Cite as: | arXiv:2407.14562 [cs.AI] |
| (or arXiv:2407.14562v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2407.14562 arXiv-issued DOI via DataCite |
Submission history
From: Yongxin Deng [view email]
[v1]
Thu, 18 Jul 2024 18:52:10 UTC (277 KB)
[v2]
Sat, 10 Aug 2024 06:54:20 UTC (277 KB)
[v3]
Thu, 8 Oct 2026 07:25:49 UTC (285 KB)
来源：arXiv：cs.CL · arxiv.org