Parallel-SFT: Improving Zero-Shot Cross-Programming-Language Transfer for Code RL
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915951554854912 |
|---|---|
| author | Wu, Zhaofeng Wang, Shiqi Peng, Boya Goyal, Anuj Kambadur, Melanie Ruder, Sebastian Kim, Yoon Bi, Chloe |
| author_facet | Wu, Zhaofeng Wang, Shiqi Peng, Boya Goyal, Anuj Kambadur, Melanie Ruder, Sebastian Kim, Yoon Bi, Chloe |
| contents | Modern language models demonstrate impressive coding capabilities in common programming languages (PLs), such as C++ and Python, but their performance in lower-resource PLs is often limited by training data availability. In principle, however, most programming skills are universal across PLs, so the capability acquired in one PL should transfer to others. In this work, we propose the task of zero-shot cross-programming-language transfer for code RL. We find that, for Llama-3.1, RL training for code generation in a source PL fails to improve, and sometimes even degrades, the performance on other target PLs. To address this, we hypothesize that effective RL transfer requires a generalizable SFT initialization before RL. We thus propose **Parallel-SFT**, an SFT strategy that incorporates "parallel programs" -- functionally equivalent code implemented in multiple PLs -- into the data mixture. We demonstrate that this improves transferability: when we subsequently perform RL on our Parallel-SFT model, we observe better generalization to unseen PLs. Analysis of the model internal representations reveals that Parallel-SFT leads to a more functionality-centric latent space, where equivalent programs across PLs are more tightly clustered, which we hypothesize to contribute to the improved transferability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_20835 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Parallel-SFT: Improving Zero-Shot Cross-Programming-Language Transfer for Code RL Wu, Zhaofeng Wang, Shiqi Peng, Boya Goyal, Anuj Kambadur, Melanie Ruder, Sebastian Kim, Yoon Bi, Chloe Computation and Language Modern language models demonstrate impressive coding capabilities in common programming languages (PLs), such as C++ and Python, but their performance in lower-resource PLs is often limited by training data availability. In principle, however, most programming skills are universal across PLs, so the capability acquired in one PL should transfer to others. In this work, we propose the task of zero-shot cross-programming-language transfer for code RL. We find that, for Llama-3.1, RL training for code generation in a source PL fails to improve, and sometimes even degrades, the performance on other target PLs. To address this, we hypothesize that effective RL transfer requires a generalizable SFT initialization before RL. We thus propose **Parallel-SFT**, an SFT strategy that incorporates "parallel programs" -- functionally equivalent code implemented in multiple PLs -- into the data mixture. We demonstrate that this improves transferability: when we subsequently perform RL on our Parallel-SFT model, we observe better generalization to unseen PLs. Analysis of the model internal representations reveals that Parallel-SFT leads to a more functionality-centric latent space, where equivalent programs across PLs are more tightly clustered, which we hypothesize to contribute to the improved transferability. |
| title | Parallel-SFT: Improving Zero-Shot Cross-Programming-Language Transfer for Code RL |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2604.20835 |