NALA_MAINZ at BLP-2025 Task 2: A Multi-agent Approach for Bangla Instruction to Python Code Generation
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_ | 1866914166096265216 |
|---|---|
| author | Saadi, Hossain Shaikh Alam, Faria Sanz-Guerrero, Mario Bui, Minh Duc Mager, Manuel von der Wense, Katharina |
| author_facet | Saadi, Hossain Shaikh Alam, Faria Sanz-Guerrero, Mario Bui, Minh Duc Mager, Manuel von der Wense, Katharina |
| contents | This paper presents JGU Mainz's winning system for the BLP-2025 Shared Task on Code Generation from Bangla Instructions. We propose a multi-agent-based pipeline. First, a code-generation agent produces an initial solution from the input instruction. The candidate program is then executed against the provided unit tests (pytest-style, assert-based). Only the failing cases are forwarded to a debugger agent, which reruns the tests, extracts error traces, and, conditioning on the error messages, the current program, and the relevant test cases, generates a revised solution. Using this approach, our submission achieved first place in the shared task with a $Pass@1$ score of 95.4. We also make our code public. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_16787 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | NALA_MAINZ at BLP-2025 Task 2: A Multi-agent Approach for Bangla Instruction to Python Code Generation Saadi, Hossain Shaikh Alam, Faria Sanz-Guerrero, Mario Bui, Minh Duc Mager, Manuel von der Wense, Katharina Computation and Language Software Engineering This paper presents JGU Mainz's winning system for the BLP-2025 Shared Task on Code Generation from Bangla Instructions. We propose a multi-agent-based pipeline. First, a code-generation agent produces an initial solution from the input instruction. The candidate program is then executed against the provided unit tests (pytest-style, assert-based). Only the failing cases are forwarded to a debugger agent, which reruns the tests, extracts error traces, and, conditioning on the error messages, the current program, and the relevant test cases, generates a revised solution. Using this approach, our submission achieved first place in the shared task with a $Pass@1$ score of 95.4. We also make our code public. |
| title | NALA_MAINZ at BLP-2025 Task 2: A Multi-agent Approach for Bangla Instruction to Python Code Generation |
| topic | Computation and Language Software Engineering |
| url | https://arxiv.org/abs/2511.16787 |