Quantified Explainability and Robustness Analysis of Transformer-Based Bug Detection Models

We are pleased to share that our research team member Debargha Ghosh has successfully published a new paper titled “Quantified Explainability and Robustness Analysis of Transformer-Based Bug Detection Models” in the International Multidisciplinary Research Journal Reviews (IMRJR).

This research investigates the application of transformer-based machine learning models in software bug detection, focusing on improving the explainability and robustness of AI-driven software analysis systems. By introducing a framework for quantified explainability and systematic evaluation, the study helps developers better understand how AI models make predictions and how reliable those predictions are in real-world software engineering environments.

The work contributes to the growing field of AI-assisted software development, where machine learning models are increasingly used to detect bugs, improve code quality, and support more efficient software maintenance processes.

We congratulate Debargha Ghosh for this excellent achievement and contribution to advancing research in AI, machine learning, and intelligent software engineering systems.

🔗 Read the full paper:

https://imrjr.com/papers/quantified-explainability-and-robustness-analysis-of-transformer-based-bug-detection-models/