影响力指数

84.07/100

前 1%

全站排名 #657

发表论文22 篇

平均评分5.7

年均产出7.3 篇/年

Han Zhao

Assistant Professor@University of Illinois, Urbana Champaign·美国·OpenReview

研究方向

trustworthy machine learning · domain adaptation; domain generalization; out-of-distribution generalization · probabilistic circuits · algorithmic fairness

4.7

Group Fairness Meets the Black Box: Enabling Fair Algorithms on Closed LLMs via Post-Processing

ICLR 2026Rejected

三作

4.5

Group Fairness Under Distribution Shifts: Analysis and Robust Post-Processing

ICLR 2026Rejected

二作

4.5

Differentially Private Conditional Text Generation with RL-Boosted Control

ICLR 2026Rejected

3.3

Task Vector Bases: A Unified and Scalable Framework for Compressed Task Arithmetic

ICLR 2026Withdrawn

通讯

8.2

Understanding and Improving Adversarial Robustness of Neural Probabilistic Circuits

NeurIPS 2025Poster

二作

8.2

A Snapshot of Influence: A Local Data Attribution Framework for Online Reinforcement Learning

NeurIPS 2025Oral

通讯

6.8

GraSS: Scalable Data Attribution with Gradient Sparsification and Sparse Projection

NeurIPS 2025Poster

6.8

Taming Hyperparameter Sensitivity in Data Attribution: Practical Selection Without Costly Retraining

NeurIPS 2025Poster

6.8

Efficient Utility-Preserving Machine Unlearning with Implicit Gradient Surgery

NeurIPS 2025Poster

通讯

6.6

Learning Structured Representations by Embedding Class Hierarchy with Fast Optimal Transport

合作者 (20)

Han Zhao

Group Fairness Meets the Black Box: Enabling Fair Algorithms on Closed LLMs via Post-Processing

Group Fairness Under Distribution Shifts: Analysis and Robust Post-Processing

Differentially Private Conditional Text Generation with RL-Boosted Control

Task Vector Bases: A Unified and Scalable Framework for Compressed Task Arithmetic

Understanding and Improving Adversarial Robustness of Neural Probabilistic Circuits

A Snapshot of Influence: A Local Data Attribution Framework for Online Reinforcement Learning

GraSS: Scalable Data Attribution with Gradient Sparsification and Sparse Projection

Taming Hyperparameter Sensitivity in Data Attribution: Practical Selection Without Costly Retraining

Efficient Utility-Preserving Machine Unlearning with Implicit Gradient Surgery

Learning Structured Representations by Embedding Class Hierarchy with Fast Optimal Transport

Accelerating Neural ODEs: A Variational Formulation-based Approach

Scaling Laws for Multilingual Language Models

Multiobjective distribution matching