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
May 2024. LATS is accepted at ICML 2024
March 2024. 1 paper is accepted at LLM Agents @ ICLR 2024
March 2024. 3 papers accepted at SeT LLM @ ICLR 2024
March 2024. LATS reaches 400 stars on GitHub and is implemented in LangChain
February 2024. 1 paper is accepted at CVPR 2024
September 2023. 2 papers accepted at NeurIPS 2023
June 2023. 1 paper is accepted at FL @ ICML 2023
June 2023. 1 paper is accepted at ICCV 2023
|
|
|
Towards Robust Unlearning in LLMs
Rishub Tamirisa, Bhrugu Bharathi, Andy Zhou, Bo Li, Mantas Mazeika
Secure and Trustworthy LLMs @ ICLR, 2024
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
Rishub Tamirisa, Chulin Xie, Wenxuan Bao, Andy Zhou, Ron Arel, Aviv Shamsian
CVPR, 2024
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
Andy Zhou, Bo Li, Haohan Wang
Secure and Trustworthy LLMs @ ICLR, 2024
arxiv /
code /
paper
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
Haibo Jin*, Ruoxi Chen*, Andy Zhou, Jinyin Chen, Yang Zhang, Haohan Wang
Secure and Trustworthy LLMs @ ICLR, 2024
arxiv /
paper
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
Andy Zhou, Kai Yan, Michal Shlapentokh Rothman, Haohan Wang, Yuxiong Wang
ICML, 2024
arxiv /
code /
website /
paper
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
Andy Zhou*, Samuel Li*, Pranav Sriram*,Xiang Li*, Jiahua Dong*, Ansh Sharma, Yuanyi Zhong, Shirui Luo, Volodymyr Kindratenko, George Heintz, Christopher Zallek, Yuxiong Wang
NeurIPS Datasets and Benchmarks, 2023
arxiv /
paper
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
Andy Zhou, Jindong Wang, Haohan Wang, Yuxiong Wang
NeurIPS, 2023
arxiv /
code /
paper
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
Rishub Tamirisa, John Won, Chengjun Lu, Ron Arel, Andy Zhou
Federated Learning @ ICML, 2023
arxiv /
paper
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
Zeyi Huang, Andy Zhou, Zijian Lin, Mu Cai, Haohan Wang, Yong Jae Lee
ICCV, 2023
arxiv /
code /
paper
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.
|
|