Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2502.12589 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912235949916160 |
|---|---|
| author | Zhang, Yu Peng, Shujun Wu, Nengwu Lin, Xinhan Hu, Yang Tang, Jie |
| author_facet | Zhang, Yu Peng, Shujun Wu, Nengwu Lin, Xinhan Hu, Yang Tang, Jie |
| contents | Recently, substantial advancements have been made in training language models to carry out step-by-step reasoning for solving intricate numerical reasoning tasks. Beyond the methods used to solve these problems, the structure and formulation of the problems themselves also play a crucial role in determining the performance of large language models. We observe that even small changes in the surface form of mathematical problems can have a profound impact on both the answer distribution and solve rate. This highlights the vulnerability of LLMs to surface-level variations, revealing its limited robustness when reasoning through complex problems. In this paper, we propose RM-PoT, a three-stage framework that integrates problem reformulation (RM), code-aided reasoning (PoT), and domain-aware few-shot learning to address these limitations. Our approach first reformulates the input problem into diverse surface forms to reduce structural bias, then retrieves five semantically aligned examples from a pre-constructed domain-specific question bank to provide contextual guidance, and finally generates executable Python code for precise computation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_12589 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RM-PoT: Reformulating Mathematical Problems and Solving via Program of Thoughts Zhang, Yu Peng, Shujun Wu, Nengwu Lin, Xinhan Hu, Yang Tang, Jie Artificial Intelligence Recently, substantial advancements have been made in training language models to carry out step-by-step reasoning for solving intricate numerical reasoning tasks. Beyond the methods used to solve these problems, the structure and formulation of the problems themselves also play a crucial role in determining the performance of large language models. We observe that even small changes in the surface form of mathematical problems can have a profound impact on both the answer distribution and solve rate. This highlights the vulnerability of LLMs to surface-level variations, revealing its limited robustness when reasoning through complex problems. In this paper, we propose RM-PoT, a three-stage framework that integrates problem reformulation (RM), code-aided reasoning (PoT), and domain-aware few-shot learning to address these limitations. Our approach first reformulates the input problem into diverse surface forms to reduce structural bias, then retrieves five semantically aligned examples from a pre-constructed domain-specific question bank to provide contextual guidance, and finally generates executable Python code for precise computation. |
| title | RM-PoT: Reformulating Mathematical Problems and Solving via Program of Thoughts |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2502.12589 |