Associated Publication

This repository contains the implementation and experimental framework for our IEEE Access publication:

Automated Resume Evaluation Using Large Language Models: A Multi-Agent Framework with Fine-Tuning and Prompt Engineering

The study investigates automated resume screening using traditional machine learning, deep learning, transformer-based models, fine-tuned large language models (LLMs), and heterogeneous multi-agent architectures. Our proposed framework employs specialized agents for Education, Skills, and Experience evaluation, each powered by independently fine-tuned language models.

The best-performing configuration (QwenFT + QwenFT + GemmaFT) achieved:

• MAE: 0.065
• Pearson Correlation (r): 0.896
• R²: 0.766

The proposed approach outperformed multiple commercial LLMs, including GPT-5, Gemini, Claude, Grok, and Mistral, while providing interpretable component-level assessments for resume evaluation.

Citation

@article{chowdhury2026resume,
  title={Automated Resume Evaluation Using Large Language Models: A Multi-Agent Framework with Fine-Tuning and Prompt Engineering},
  author={Chowdhury, Md. Sagor and Chowdhury, Adiba Fairooz and Banu, Ayesha and Hossain, Riad and Chowdhury, Mahfuzulhoq},
  journal={IEEE Access},
  year={2026},
  doi={10.1109/ACCESS.2026.3696456}
}

If you use this repository in your research, please cite the above paper.