Explainable AI

Goal: Making AI systems understandable, transparent, and accountable.

Research Questions

Why did a model make a particular decision?

What internal mechanisms drive model behavior?

How can explanations help users trust AI systems?

Representative Publications

A Counterfactual Explanation Framework for Retrieval Models (ACL 2026)

Dissecting Bias in LLMs: A Mechanistic Interpretability Perspective (TMLR 2025)

The Curious Case of IR Explainability: Explaining Document Scores within and across Ranking Models (SIGIR 2020)

Measuring and comparing the consistency of IR models for query pairs with similar and different information needs (CIKM 2022)