Enregistré dans:
Détails bibliographiques
Auteurs principaux: Yu, Yiyao, Zhang, Yuxiang, Zhang, Dongdong, Liang, Xiao, Zhang, Hengyuan, Zhang, Xingxing, Yang, Ziyi, Khademi, Mahmoud, Awadalla, Hany, Wang, Junjie, Yang, Yujiu, Wei, Furu
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2501.11110
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908518717587456
author Yu, Yiyao
Zhang, Yuxiang
Zhang, Dongdong
Liang, Xiao
Zhang, Hengyuan
Zhang, Xingxing
Yang, Ziyi
Khademi, Mahmoud
Awadalla, Hany
Wang, Junjie
Yang, Yujiu
Wei, Furu
author_facet Yu, Yiyao
Zhang, Yuxiang
Zhang, Dongdong
Liang, Xiao
Zhang, Hengyuan
Zhang, Xingxing
Yang, Ziyi
Khademi, Mahmoud
Awadalla, Hany
Wang, Junjie
Yang, Yujiu
Wei, Furu
contents Large Language Models (LLMs) have made notable progress in mathematical reasoning, yet often rely on single-paradigm reasoning, limiting their effectiveness across diverse tasks. We introduce Chain-of-Reasoning (CoR), a novel unified framework integrating multiple reasoning paradigms--Natural Language Reasoning (NLR), Algorithmic Reasoning (AR), and Symbolic Reasoning (SR)--to enable synergistic collaboration. CoR generates multiple potential answers via different reasoning paradigms and synthesizes them into a coherent final solution. We propose a Progressive Paradigm Training (PPT) strategy for models to progressively master these paradigms, leading to CoR-Math-7B. Experimental results demonstrate that CoR-Math-7B significantly outperforms current SOTA models, achieving up to a 41.0% absolute improvement over GPT-4o in theorem proving and a 15.0% improvement over RL-based methods on the MATH benchmark in arithmetic tasks. These results show the enhanced mathematical comprehension ability of our model, enabling zero-shot generalization across tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2501_11110
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Chain-of-Reasoning: Towards Unified Mathematical Reasoning in Large Language Models via a Multi-Paradigm Perspective
Yu, Yiyao
Zhang, Yuxiang
Zhang, Dongdong
Liang, Xiao
Zhang, Hengyuan
Zhang, Xingxing
Yang, Ziyi
Khademi, Mahmoud
Awadalla, Hany
Wang, Junjie
Yang, Yujiu
Wei, Furu
Computation and Language
Large Language Models (LLMs) have made notable progress in mathematical reasoning, yet often rely on single-paradigm reasoning, limiting their effectiveness across diverse tasks. We introduce Chain-of-Reasoning (CoR), a novel unified framework integrating multiple reasoning paradigms--Natural Language Reasoning (NLR), Algorithmic Reasoning (AR), and Symbolic Reasoning (SR)--to enable synergistic collaboration. CoR generates multiple potential answers via different reasoning paradigms and synthesizes them into a coherent final solution. We propose a Progressive Paradigm Training (PPT) strategy for models to progressively master these paradigms, leading to CoR-Math-7B. Experimental results demonstrate that CoR-Math-7B significantly outperforms current SOTA models, achieving up to a 41.0% absolute improvement over GPT-4o in theorem proving and a 15.0% improvement over RL-based methods on the MATH benchmark in arithmetic tasks. These results show the enhanced mathematical comprehension ability of our model, enabling zero-shot generalization across tasks.
title Chain-of-Reasoning: Towards Unified Mathematical Reasoning in Large Language Models via a Multi-Paradigm Perspective
topic Computation and Language
url https://arxiv.org/abs/2501.11110