Andy Zhou

I am a BS/MS student at the University of Illinois at Urbana-Champaign advised by Bo Li. I also work closely with Yuxiong Wang and Haohan Wang.

My research interests are in trustworthy machine learning, AI safety, and LM agents. My work focuses on improving the capabilities and reliability of large language models.

I am also the founder and Head of Research at Lapis Labs, a student-led research group.

Email  /  GitHub  /  Google Scholar  /  Twitter

News

Research

Towards Robust Unlearning in LLMs

We outline the setting of robust machine unlearning in LLMs for reliably removing unwanted knowledge.

FedSelect: Personalized Federated Learning with Customized Selection of Parameters for Fine-Tuning

We propose a federated-learning algorithm based on selecting which parameters to use for fine-tuning and which to make global updates.

Robust Prompt Optimization for Defending Language Models Against Jailbreaking Attacks

We propose a defense objective for defending LLMs against jailbreaking and an algorithm to generate trigger tokens that enforce harmless behavior, improving robustness across jailbreaks and models.

GUARD: Role-playing to Generate Natural-language Jailbreakings to Test Guideline Adherence of Large Language Models

We propose a framework to generate semantic jailbreaks from human safety guidelines using syntatic parsing organized into knowledge graphs and LM optimization. Jailbreaks are SOTA for success rate and work on VLMS.

Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models

We propose the first search algorithm for LM agents which draws upon aspects of reasoning and acting prompting methods to improve decision-making. We achieve SOTA on HumanEval with a Pass@1 rate of 94.4%

YouTubePD: A Multimodal Benchmark for Parkinson’s Disease Analysis

We propose the first public benchmark for automated Parkinson’s disease analysis. We explore three tasks–facial-expression-based PD classification, multimodal PD classification, and PD progression synthesis–and show models trained on YouTubePD generalize to real clinical data.

Distilling Out-of-Distribution Robustness from Vision-Language Foundation Models

We propose a data augmentation and knowledge distillation objective that uses teacher gradients to generate diverse samples, improving out-of-distribution robustness. We distill from CLIP to train the most robust ResNet34 and ResNet50 on OOD generalization.

FedSelect: Customized Selection of Parameters for Fine-Tuning during Personalized Federated Learning

We propose a federated-learning algorithm based on selecting which parameters to use for fine-tuning and which to make global updates.

A Sentence Speaks a Thousand Images: Domain Generalization through Distilling CLIP with Language Guidance

We propose a distillation objective based on CLIP text representations to improve domain generalization.

Misc

My sister Anna is an Assistant Professor of Pharmacy at Loma Linda University.

Design and source code from Jon Barron's website