PEEM paper published in IEEE Access! 🏆

I am thrilled to announce that our paper, “PEEM: Prompt Engineering Evaluation Metrics for Interpretable Joint Evaluation of Prompts and Responses,” has been officially published in IEEE Access! 📄

This research introduces a novel evaluation framework, PEEM, designed to jointly assess the quality of prompts and the corresponding model responses. Unlike traditional metrics that often treat the model as a black box, PEEM provides interpretable insights into the interaction between prompt engineering and output performance.

💡 Key Contributions

1️⃣ Joint Evaluation Framework: We move beyond simple response scoring by simultaneously evaluating the instructional quality of prompts and the resulting response accuracy.

2️⃣ Interpretable Metrics: PEEM offers granular metrics that help researchers and practitioners understand why a model succeeds or fails based on specific prompt characteristics.

3️⃣ Extensive Validation: We demonstrated PEEM’s effectiveness through rigorous testing across multiple large language models (LLMs) and diverse prompt engineering strategies.

You can access the full paper through the following links:

• IEEE Xplore: https://ieeexplore.ieee.org/document/11460151
• arXiv (Preprint): https://arxiv.org/abs/2603.10477

I’m incredibly grateful to my co-authors and the reviewers for their valuable feedback and support throughout this process! 🙏