Towards Better Understanding of Program-of-Thought Reasoning in Cross-Lingual and Multilingual Environments
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909620222558208 |
|---|---|
| author | Payoungkhamdee, Patomporn Tuchinda, Pume Baek, Jinheon Cahyawijaya, Samuel Udomcharoenchaikit, Can Manakul, Potsawee Limkonchotiwat, Peerat Chuangsuwanich, Ekapol Nutanong, Sarana |
| author_facet | Payoungkhamdee, Patomporn Tuchinda, Pume Baek, Jinheon Cahyawijaya, Samuel Udomcharoenchaikit, Can Manakul, Potsawee Limkonchotiwat, Peerat Chuangsuwanich, Ekapol Nutanong, Sarana |
| contents | Multi-step reasoning is essential for large language models (LLMs), yet multilingual performance remains challenging. While Chain-of-Thought (CoT) prompting improves reasoning, it struggles with non-English languages due to the entanglement of reasoning and execution. Program-of-Thought (PoT) prompting separates reasoning from execution, offering a promising alternative but shifting the challenge to generating programs from non-English questions. We propose a framework to evaluate PoT by separating multilingual reasoning from code execution to examine (i) the impact of fine-tuning on question-reasoning alignment and (ii) how reasoning quality affects answer correctness. Our findings demonstrate that PoT fine-tuning substantially enhances multilingual reasoning, outperforming CoT fine-tuned models. We further demonstrate a strong correlation between reasoning quality (measured through code quality) and answer accuracy, highlighting its potential as a test-time performance improvement heuristic. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_17956 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Towards Better Understanding of Program-of-Thought Reasoning in Cross-Lingual and Multilingual Environments Payoungkhamdee, Patomporn Tuchinda, Pume Baek, Jinheon Cahyawijaya, Samuel Udomcharoenchaikit, Can Manakul, Potsawee Limkonchotiwat, Peerat Chuangsuwanich, Ekapol Nutanong, Sarana Computation and Language Multi-step reasoning is essential for large language models (LLMs), yet multilingual performance remains challenging. While Chain-of-Thought (CoT) prompting improves reasoning, it struggles with non-English languages due to the entanglement of reasoning and execution. Program-of-Thought (PoT) prompting separates reasoning from execution, offering a promising alternative but shifting the challenge to generating programs from non-English questions. We propose a framework to evaluate PoT by separating multilingual reasoning from code execution to examine (i) the impact of fine-tuning on question-reasoning alignment and (ii) how reasoning quality affects answer correctness. Our findings demonstrate that PoT fine-tuning substantially enhances multilingual reasoning, outperforming CoT fine-tuned models. We further demonstrate a strong correlation between reasoning quality (measured through code quality) and answer accuracy, highlighting its potential as a test-time performance improvement heuristic. |
| title | Towards Better Understanding of Program-of-Thought Reasoning in Cross-Lingual and Multilingual Environments |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2502.17956 |