Adversarial Attacks on Post Hoc Explanation Techniques (LIME/SHAP)

Reinforced Causal Explainer for Graph Neural Networks, TPAMI2022

Benchmark to Evaluate EXplainable AI

An eXplainable AI system to elucidate short-term speed forecasts in traffic networks obtained by Spatio-Temporal Graph Neural Networks.

[ACM MM 2024] Holistic-CAM: Ultra-lucid and Sanity Preserving Visual Interpretation in Holistic Stage of CNNs

[SIGIR 2025] Class Activation Values: Lucid and Faithful Visual Interpretations for CLIP-based Text-Image Retrievals.

Human-centered XAI via a Concept-Informed Prompt-based Validation framework for saliency maps [CIProVa]

[AAAI'23 Paper] A machine learning defense for auditors of black box automated decision-making systems.

CAM, Grad-CAM, Grad-CAM++ and Guided Backpropagation post-hoc explanation methods

Applying post-hoc attention with CAM to the MNIST dataset

Astrapia - A Friendly XAI Explainer Evaluation Framework

This repository presents a comprehensive research paper exploring the role of Explainable Artificial Intelligence (XAI) in modern Machine Learning. It aims to shed light on the interpretability of 'black-box' models like Neural Networks, Explainable AI and highlights the need for transparent, human-understandable ML systems.

Repository containing code for Black-box Association-Rule Based Explanations (BARBE). Main code is in the barbe directory, see the Readme below for more details.