Nash CoT: Multi-Path Inference with Preference Equilibrium

Ziqi Zhang, Cunxiang Wang, Xiao Xiong, Yue Zhang, Donglin Wang


Abstract
Chain of thought (CoT) is a reasoning framework that can enhance the performance of large language models (LLMs) on complex inference tasks. In particular, among various studies related to CoT, multi-path inference stands out as a simple yet effective improvement. However, there is no optimal setting for the number of inference paths. Therefore, we have to increase the number of inference paths to obtain better results, which in turn increases the inference cost. To address this limitation, we can utilize question-related role templates to guide LLMs into relevant roles, thereby increasing the possibility of correct inferences for each path and further reducing dependence on the number of inference paths while improving reasoning accuracy. However, placing LLMs into specific roles may reduce their reasoning diversity and performance on a few tasks where role dependence is low. To alleviate the excessive immersion of the LLM into a specific role, we propose Nash CoT by constructing a competitive system on each path that balances the generation from role-specific LLMs’ and the general LLMs’ generation, thereby ensuring both effective role adoption and diversity in LLM generation further maintaining the performance of multi-path inference while reducing the requirement of the number of inference paths. We evaluate Nash CoT across various inference tasks, including Arabic Reasoning, Commonsense Question Answering, and Symbolic Inference, achieving results that are comparable to or better than those of multi-path CoT with the equal number of inference paths.
Anthology ID:
2024.emnlp-main.807
Volume:
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
14572–14587
Language:
URL:
https://aclanthology.org/2024.emnlp-main.807
DOI:
Bibkey:
Cite (ACL):
Ziqi Zhang, Cunxiang Wang, Xiao Xiong, Yue Zhang, and Donglin Wang. 2024. Nash CoT: Multi-Path Inference with Preference Equilibrium. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 14572–14587, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
Nash CoT: Multi-Path Inference with Preference Equilibrium (Zhang et al., EMNLP 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.emnlp-main.807.pdf
Software:
 2024.emnlp-main.807.software.zip