Enhancing Mathematical Problem Solving in LLMs through Execution-Driven Reasoning Augmentation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917256549629952 |
|---|---|
| author | Basarkar, Aditya Tabarsi, Benyamin Barnes, Tiffany Xu, Dongkuan |
| author_facet | Basarkar, Aditya Tabarsi, Benyamin Barnes, Tiffany Xu, Dongkuan |
| contents | Mathematical problem solving is a fundamental benchmark for assessing the reasoning capabilities of artificial intelligence and a gateway to applications in education, science, and engineering where reliable symbolic reasoning is essential. Although recent advances in multi-agent LLM-based systems have enhanced their mathematical reasoning capabilities, they still lack a reliably revisable representation of the reasoning process. Existing agents either operate in rigid sequential pipelines that cannot correct earlier steps or rely on heuristic self-evaluation that can fail to identify and fix errors. In addition, programmatic context can distract language models and degrade accuracy. To address these gaps, we introduce Iteratively Improved Program Construction (IIPC), a reasoning method that iteratively refines programmatic reasoning chains and combines execution feedback with the native Chain-of-thought abilities of the base LLM to maintain high-level contextual focus. IIPC surpasses competing approaches in the majority of reasoning benchmarks on multiple base LLMs. All code and implementations are released as open source. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_03950 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Enhancing Mathematical Problem Solving in LLMs through Execution-Driven Reasoning Augmentation Basarkar, Aditya Tabarsi, Benyamin Barnes, Tiffany Xu, Dongkuan Artificial Intelligence Machine Learning Multiagent Systems 68T20, 68T50 I.2.7; I.2.11; I.2.1 Mathematical problem solving is a fundamental benchmark for assessing the reasoning capabilities of artificial intelligence and a gateway to applications in education, science, and engineering where reliable symbolic reasoning is essential. Although recent advances in multi-agent LLM-based systems have enhanced their mathematical reasoning capabilities, they still lack a reliably revisable representation of the reasoning process. Existing agents either operate in rigid sequential pipelines that cannot correct earlier steps or rely on heuristic self-evaluation that can fail to identify and fix errors. In addition, programmatic context can distract language models and degrade accuracy. To address these gaps, we introduce Iteratively Improved Program Construction (IIPC), a reasoning method that iteratively refines programmatic reasoning chains and combines execution feedback with the native Chain-of-thought abilities of the base LLM to maintain high-level contextual focus. IIPC surpasses competing approaches in the majority of reasoning benchmarks on multiple base LLMs. All code and implementations are released as open source. |
| title | Enhancing Mathematical Problem Solving in LLMs through Execution-Driven Reasoning Augmentation |
| topic | Artificial Intelligence Machine Learning Multiagent Systems 68T20, 68T50 I.2.7; I.2.11; I.2.1 |
| url | https://arxiv.org/abs/2602.03950 |