Sifat Hossain

Publications

* denotes equal contribution. Reverse-chronological order. Click Abstract or BibTeX on any entry to expand.

Google Scholar·15 citations·h-index 1
  1. [1]
    A-ProS framework diagram: problem acquisition and preprocessing feeds a GPT-based solution generator, whose outputs are submitted to Codeforces; verdicts drive a multi-model debugging feedback loop (Codestral, Llama-3.3, DeepSeek-R1) that iteratively refines the solution.

    A-ProS: Towards Reliable Autonomous Programming Through Multi-Model Feedback

    Anika Tabassum*, Md Sifat Hossain*, Md. Fahim Arefin, Tariqul Islam, Tarannum Shaila Zaman

    ACM Transactions on Software Engineering and Methodology (TOSEM), 2026 · * Equal contribution

    TOSEM 2026arXivPDFProject PageCode
  2. [2]
    LLM-ProS pipeline: ICPC problem data collection, data preprocessing, model testing across GPT-4o / Mistral Large / Llama / OpenAI o1 models, and automated submission with verdict classification (Accepted, Compilation error, Time limit exceeded, Wrong answer).

    LLM-ProS: Analyzing Large Language Models’ Performance in Competitive Problem Solving

    Md Sifat Hossain, Anika Tabassum, Md. Fahim Arefin, Tarannum Shaila Zaman

    LLM4Code 2025 Workshop, ICSE 2025 - 47th International Conference on Software Engineering, Ottawa, Canada

    ICSE 2025 (LLM4Code)arXivPDFCode