(WWW'21) ATON - an Outlier Interpreation / Outlier explanation method

Codebase for "Demystifying Black-box Models with Symbolic Metamodels", NeurIPS 2019.

We introduce XBrainLab, an open-source user-friendly software, for accelerated interpretation of neural patterns from EEG data based on cutting-edge computational approach.

CAVES-dataset accepted at SIGIR'22

[TMLR] "Can You Win Everything with Lottery Ticket?" by Tianlong Chen, Zhenyu Zhang, Jun Wu, Randy Huang, Sijia Liu, Shiyu Chang, Zhangyang Wang

Comprehensible Convolutional Neural Networks via Guided Concept Learning

A solution to the Boolean Blindness problem.

tornado plots for model sensitivity analysis

MechaMap - Toolkit for Mechanistic Interpretability (MI) Research

List of papers in the area of Explainable Artificial Intelligence Year wise

Code for ER-Test, accepted to the Findings of EMNLP 2022

A framework for evaluating auto-interp pipelines, i.e., natural language explanations of neurons.

The mechanisms behind image classification using a pretrained CNN model in high-dimensional spaces 🏞️

Experiments to explain entity resolution systems

Code for actual causes filtering based on relevance metrics.

Interpretability analysis of Vision Transformers using Attention Rollout, Gradient Rollout, and a novel FAF/BAF metric with Mask R-CNN. Includes experiments on DeiT-Tiny comparing CNN and ViT explanation behavior, showing ~25% improvement in semantic focus using gradient-based rollout.

ML Pipeline. Detail documentation of the project in README. Click on actions to see the script.

A project in an AI seminar

Demo of loan-approval prediction with per-feature SHAP explanations and optional interest-rate recommendations; includes FastAPI inference and a Streamlit UI.

Domestic robot example configured for the multi-level explainability framework

Quantifying the robustness of XAI in financial fraud detection. A systematic verification framework for assessing the stability and fidelity of additive local explanations (SHAP/LIME) under controlled perturbations.