Chejian Xu

Chejian Xu

I am a third year Computer Science Ph.D. student at University of Illinois, Urbana-Champaign (UIUC), advised by Prof. Bo Li. I received my Bachelor's degree from Computer Science, Zhejiang University at CKC Honors College, advised by Prof. Shouling Ji and Prof. Siliang Tang.

My current research interests are focused on advancing the security, robustness, and generalization of machine learning (ML) systems. My work delves into the intriguing intersection of these critical aspects, particularly within the domains of natural language processing (NLP) and reinforcement learning (RL). My primary goal is to develop innovative methodologies and techniques that enhance the reliability and trustworthiness of ML models, such as language models (LMs), within these domains.

Email / Google Scholar / Github / LinkedIn

News

2024/09 - We released AdvWeb, a controllable black-box attack on VLM-powered web agents.
2024/09 - We released MMDT, providing comprehensive assessment of trustworthiness in multimodal foundation models.
2024/09 - We released ChatQA 2, a Llama 3.0-based model with enhanced long-context understanding and RAG capabilities.
2024/09 - Our paper, DecodingTrust, got the Cybersecurity award 2024 on Best Machine Learning and Security Paper.
2024/05 - I started my internship at NVIDIA, working on long context LLMs.
2024/04 - We are hosting the The Competition for LLM and Agent Safety 2024!
2024/02 - One paper got accepted to CVPR 2024.
2023/12 - Our paper, DecodingTrust, received the Outstanding Paper award at NeurIPS 2023.
2023/09 - One paper got accepted to NeurIPS 2023.
2023/03 - We are hosting the Secure and Safe Autonomous Driving (SSAD) Workshop and Challenge at CVPR 2023!

Publications

	AdvWeb: Controllable Black-box Attacks on VLM-powered Web Agents Chejian Xu, Mintong Kang, Jiawei Zhang, Zeyi Liao, Lingbo Mo, Mengqi Yuan, Huan Sun, Bo Li Preprint, 2024 [PDF] [Code] [Website] [BibTeX]
	MMDT: Decoding the Trustworthiness and Safety of Multimodal Foundation Models Chejian Xu, Jiawei Zhang, Zhaorun Chen, Chulin Xie, Mintong Kang, Zhuowen Yuan, Zidi Xiong, Chenhui Zhang, Lingzhi Yuan, Yi Zeng, Peiyang Xu, Chengquan Guo, Andy Zhou, Jeffrey Ziwei Tan, Zhun Wang, Alexander Xiong, Xuandong Zhao, Yu Gai, Francesco Pinto, Yujin Potter, Zhen Xiang, Zinan Lin, Dan Hendrycks, Dawn Song, Bo Li Preprint, 2024 [Code] [Website] [T2I Dataset 🤗] [I2T Dataset 🤗]
	EIA: Environmental Injection Attack on Generalist Web Agents for Privacy Leakage Zeyi Liao, Lingbo Mo, Chejian Xu, Mintong Kang, Jiawei Zhang, Chaowei Xiao, Yuan Tian, Bo Li, Huan Sun Preprint, 2024 [PDF] [Code] [BibTeX]
	ChatQA 2: Bridging the Gap to Proprietary LLMs in Long Context and RAG Capabilities Peng Xu, Wei Ping, Xianchao Wu, Chejian Xu, Zihan Liu, Mohammad Shoeybi, Bryan Catanzaro Preprint, 2024 [PDF] [Website] [Model Weights 🤗] [Training Data 🤗] [BibTeX]
	KnowHalu: Hallucination Detection via Multi-Form Knowledge Based Factual Checking Jiawei Zhang, Chejian Xu, Yu Gai, Freddy Lecue, Dawn Song, Bo Li Preprint, 2024 [PDF] [Code] [BibTeX]
	ChatScene: Knowledge-Enabled Safety-Critical Scenario Generation for Autonomous Vehicles Jiawei Zhang, Chejian Xu, Bo Li The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024 [PDF] [Code] [Website] [BibTeX]
	DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models Boxin Wang, Weixin Chen, Hengzhi Pei, Chulin Xie, Mintong Kang, Chenhui Zhang, Chejian Xu, Zidi Xiong, Ritik Dutta, Rylan Schaeffer, Sang T. Truong, Simran Arora, Mantas Mazeika, Dan Hendrycks, Zinan Lin, Yu Cheng, Sanmi Koyejo, Dawn Song, Bo Li Thirty-seventh Conference on Neural Information Processing Systems (NeurIPS), 2023 (Outstanding Paper) [PDF] [Code] [Website] [BibTeX]
	DiffScene: Diffusion-Based Safety-Critical Scenario Generation for Autonomous Vehicles Chejian Xu, Ding Zhao, Alberto Sangiovanni-Vincentelli, Bo Li Workshop on New Frontiers in Adversarial Machine Learning at ICML* 2023* [Website] [BibTeX]
	SafeBench: A Benchmarking Platform for Safety Evaluation of Autonomous Vehicles Chejian Xu, Wenhao Ding, Weijie Lyu, Zuxin Liu, Shuai Wang, Yihan He, Hanjiang Hu, Ding Zhao, Bo Li Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS), 2022 [PDF] [Code] [Leaderboard] [BibTeX]
	A Survey on Safety-Critical Driving Scenario Generation -- A Methodological Perspective Wenhao Ding, Chejian Xu, Mansur Arief, Haohong Lin, Bo Li, Ding Zhao IEEE Transactions on Intelligent Transportation Systems (T-ITS), March, 2023 [PDF] [BibTeX]
	SemAttack: Natural Textual Attacks via Different Semantic Spaces Boxin Wang, Chejian Xu, Xiangyu Liu, Yu Cheng, Bo Li North American Chapter of the Association for Computational Linguistics (NAACL), 2022 (Findings) [PDF] [Code] [BibTeX]
	Copy Motion From One to Another: Fake Motion Video Generation Zhenguang Liu, Sifan Wu, Chejian Xu, Xiang Wang, Lei Zhu, Shuang Wu, Fuli Feng 31st International Joint Conference on Artificial Intelligence (IJCAI), 2022 [PDF] [Code] [BibTeX]
	COPA: Certifying Robust Policies for Offline Reinforcement Learning against Poisoning Attacks Fan Wu, Linyi Li, Chejian Xu, Huan Zhang, Bhavya Kailkhura, Krishnaram Kenthapadi, Ding Zhao, Bo Li The Tenth International Conference on Learning Representations (ICLR), 2022 [PDF] [Code] [Leaderboard] [BibTeX]
	Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models Boxin Wang, Chejian Xu, Shuohang Wang, Zhe Gan, Yu Cheng, Jianfeng Gao, Ahmed Hassan Awadallah, Bo Li Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS), 2021 (Oral) [PDF] [Leaderboard] [Dataset] [BibTeX]

Service

Conference Reviewer: NeurIPS 2022-2024, ICLR 2025, AISTATS 2025, AAAI 2023-2024, AACL 2022
Journal Reviewer: IEEE T-ITS
Organizer: The Competition for LLM and Agent Safety 2024, CVPR 2023 SSAD Workshop and Challenge, NeurIPS 2022 DMLW Workshop
Program Committee: ICLR 2023 RTML Workshop