Shichang (Ray) Zhang
What's New
[April 2025] Our paper on A Mechanistic View of How Post-Training Reshapes LLMs has won the NENLP 2025 Outstanding Paper Award. [PDF][slides]
[April 2025] Our paper on A Mechanistic View of How Post-Training Reshapes LLMs has been selected as Oral presentation for NENLP 2025.
[April 2025] Give a talk on AI Interpretability at Georgia Institute of Technology.
[April 2025] Give a talk on AI Interpretability at Emory University.
[Feb 2025] Serving as an Area Chair for ACL ARR 2025.
[Feb 2025] Our paper on Advancing Interpretability by Unifying Feature, Data and Model Component Attribution is on arXiv now. [PDF]
[Dec 2024] Selected as the top 10% of reviewers for KDD 2025.
[Nov 2024] Our paper on LLM-based explainable molecular concept learning is accepted by COLING 2025. [PDF]
[Oct 2024] Our paper on explainable graph learning for predicting ICU length of stay is accepted by ISR. [PDF]
[Oct 2024] Our paper on generalized group data attribution is on arXiv now. [PDF]
[July 2024] A Mind Map of Knowledge in LLMs.
[June 2024] Our paper on controlling LLM behaviors with LLM itself as a judge is on arXiv now. [PDF] [Code] [Website]
[May 2024] Defended my Ph.D. thesis “Explainable AI for Graph Data”.
[May 2024] Our paper on efficient ensembling for training data attribution is on arXiv now. [PDF]
[May 2024] Our paper on measure-theoretic compact fuzzy set representation for taxonomy expansion is accepted by ACL 2024 as Findings. [PDF]
[May 2024] Two papers on GNNs for explainable material science and benchmarking LLMs on scientific problems are accepted by ICML 2024. [PDF1] [PDF2]
Find out older news
Research Interests
My research interests include
Selected Publications and Pre-prints
How Post-Training Reshapes LLMs: A Mechanistic View on Knowledge, Truthfulness, Refusal, and Confidence
Hongzhe Du*, Weikai Li*, Min Cai, Karim Saraipour, Zimin Zhang, Himabindu Lakkaraju, Yizhou Sun, Shichang Zhang (*equal contribution)
NENLP 2025 Outstanding Paper [PDF] [slides]
Building Bridges, Not Walls - Advancing Interpretability by Unifying Feature, Data and Model Component Attribution
Shichang Zhang, Tessa Han, Usha Bhalla, Himabindu Lakkaraju
BuildingTrust@ICLR 2025 [PDF]
Automated Molecular Concept Generation and Labeling with Large Language Models
Zimin Zhang*, Qianli Wu*, Botao Xia*, Fang Sun, Ziniu Hu, Yizhou Sun, Shichang Zhang (*equal contribution)
COLING 2025 [PDF] [Code]
An Explainable AI Approach using Graph Learning to Predict ICU Length of Stay
Tianjian Guo, Indranil Bardhan, Ying Ding, Shichang Zhang
ISR Oct. 2024 [PDF]
Generalized Group Data Attribution
Dan Ley, Suraj Srinivas, Shichang Zhang, Gili Rusak, Himabindu Lakkaraju
ATTRIB@NeurIPS 2024 [PDF]
Efficient Ensembles Improve Training Data Attribution
Junwei Deng*, Ting-Wei Li*, Shichang Zhang, Jiaqi Ma (*equal contribution)
DMLR@ICML 2024 [PDF] [Code]
Predicting and Interpreting Energy Barriers of Metallic Glasses with Graph Neural Networks
Haoyu Li*, Shichang Zhang*, Longwen Tang, Mathieu Bauchy, Yizhou Sun (*equal contribution)
ICML 2024 [PDF] [Code]
PaGE-Link: Graph Neural Network Explanation for Heterogeneous Link Prediction
Shichang Zhang, Jiani Zhang, Xiang Song, Soji Adeshina, Da Zheng, Christos Faloutsos, Yizhou Sun
WWW 2023 [PDF] [Code]
GStarX: Explaining Graph Neural Networks with Structure-Aware Cooperative Games
Shichang Zhang, Neil Shah, Yozen Liu, Yizhou Sun
NeurIPS 2022 [PDF] [Code]
Graph-less Neural Networks, Teach Old MLPs New Tricks via Distillation
Shichang Zhang, Yozen Liu, Yizhou Sun, Neil Shah
ICLR 2022 [PDF] [Code]
Full list of publications
Honors and Awards
|