{"ID":2839256,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.16787","arxiv_id":"2511.16787","title":"NALA_MAINZ at BLP-2025 Task 2: A Multi-agent Approach for Bangla Instruction to Python Code Generation","abstract":"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.","short_abstract":"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...","url_abs":"https://arxiv.org/abs/2511.16787","url_pdf":"https://arxiv.org/pdf/2511.16787v1","authors":"[\"Hossain Shaikh Saadi\",\"Faria Alam\",\"Mario Sanz-Guerrero\",\"Minh Duc Bui\",\"Manuel Mager\",\"Katharina von der Wense\"]","published":"2025-11-20T20:26:28Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.SE\"]","methods":"[]","has_code":false}