@inproceedings{juboraj-etal-2025-bracu,
    title = "{BRACU}{\_}{CL} at {BLP}-2025 Task 2: {C}ode{M}ist: A Transformer-Based Framework for {B}angla Instruction-to-Code Generation",
    author = "Juboraj, Md. Fahmid-Ul-Alam  and
      Niloy, Soumik Deb  and
      E Sobhani, Mahbub  and
      Sadeque, Farig",
    editor = "Alam, Firoj  and
      Kar, Sudipta  and
      Chowdhury, Shammur Absar  and
      Hassan, Naeemul  and
      Prince, Enamul Hoque  and
      Tasnim, Mohiuddin  and
      Rony, Md Rashad Al Hasan  and
      Rahman, Md Tahmid Rahman",
    booktitle = "Proceedings of the Second Workshop on Bangla Language Processing (BLP-2025)",
    month = dec,
    year = "2025",
    address = "Mumbai, India",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-ijcnlp-aacl/2025.banglalp-1.67/",
    pages = "656--662",
    ISBN = "979-8-89176-314-2",
    abstract = "This study proposes a hybrid framework for Bangla-to-Python code generation, emphasizing improved code accuracy through a two-phase pipeline: generation and debugging. During development, standalone models such as TigerLLM and StarCoder achieved modest accuracies of 27{\%} and 24{\%}, respectively, while more advanced models, Gemini-1.5-flash and Gemma, reached 60{\%} and 64{\%}. Integrating Gemma with the gpt-oss debugger substantially increased accuracy to 99.75{\%}, highlighting the critical role of a dedicated debugging stage. In testing on unseen data, gpt-oss alone achieved 67{\%}, which improved to 71{\%} with self-debugging. The highest performance, 84{\%}, was obtained by pairing Gemini-2.5-flash as the generator with gpt-oss for debugging. These findings demonstrate that combining a strong generative model with an effective debugging component yields superior and robust code generation results, outperforming existing approaches such as TigerLLM. The full implementation of the framework is publicly available at https://github.com/fahmid-juboraj/Code{\_}generation."
}