Yuji Zhang (张雨季) Postdoc University of Illinois at Urbana-Champaign Email: yujiz@illinois.edu Google Scholar Twitter

I am a postdoc researcher at the University of Illinois at Urbana-Champaign (UIUC). I am fortunately supervised by Prof. Heng Ji and Prof. Chengxiang Zhai.

Research

I have broad interests in Natural Language Processing (NLP), theoretical interpretation of large language models (LLMs), and trustworthy LLMs. I study the knowledge mechanisms of LLMs—how they acquire, store, represent, and utilize knowledge, and leverage these insights to enhance model reliability and performance. Especially, I focus on the following directions:

• Interpreting, predicting, and preventing hallucination through the lens of knowledge interaction (e.g., knowledge overshadowing).
• Updating knowledge in ways that preserve model robustness and reliability.
• Improving knowledge acquisition mechanisms to enhance model intelligence.

Recent News

Invited Talks, Tutorials, Workshops, and Sevice

• [Jan 2026] Invited Talk of "Knowledge is Power, But Power Casts Shadows?" at University of Edinburgh
• [Jan 2026] Invited Talk of "Knowledge is Power, But Power Casts Shadows?" at University of Massachusetts
• [Nov 2025] Invited Talk of "Developing Robust and Trustworthy Foundation Models" at NICE Academic Platform
• [Oct 2025] Invited Talk of "Developing Robust and Trustworthy Language Models" at Northeastern University
• [Aug 2025] Organized a Workshop on "Towards Knowledgeable Foundation Models" at ACL 2025
• [Aug 2025] Session Chair for "Language Models and Interpretability" at ACL 2025
• [Aug 2025] Session Chair for "Language Modeling" at ACL 2025
• [Apr 2025] Invited Talk of "The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination" at Ploutos
• [Apr 2025] Invited Talk of "The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination" at Chinese Academy of Sciences
• [Mar 2025] Invited Talk of "The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination" at University of Texas at Austin
• [Feb 2025] Organized a Tutorial on "The Lifecycle of Knowledge in Large Language Models: Memorization, Editing, and Beyond" at AAAI 2025
• [Aug 2024] Invited Talk of "Knowledge Overshadowing Causes Amalgamated Hallucination in Large Language Models" at Beijing Academy of Artificial Intelligence

Publications

(see full list in Google Scholar. "†" for corresponding authors.)

• 

  Atomic Reasoning for Scientific Table Claim Verification

  Yuji Zhang, Qingyun Wang, Cheng Qian, Jiateng Liu, Chenkai Sun, Denghui Zhang, Tarek Abdelzaher, Chengxiang Zhai, Preslav Nakov, Heng Ji.
  [Arxiv]

• 

  The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination*

  Yuji Zhang, Sha Li, Cheng Qian, Jiateng Liu, Pengfei Yu, Yi R. Fung, Chi Han, Kathleen McKeown, Chengxiang Zhai, Manling Li, Heng Ji.
  [ACL 2025]

• 

  Knowledge Overshadowing Causes Amalgamated Hallucination in Large Language Models*

  Yuji Zhang, Sha Li, Jiateng Liu, Pengfei Yu, Yi R. Fung, Jing Li, Manling Li, Heng Ji.
  [Arxiv]

• 

  Geometric-disentanglement Unlearning

  Duo Zhou*, Yuji Zhang*, Tianxin Wei, Ruizhong Qiu, Ke Yang, Xiao Lin, Cheng Qian,

  Jingrui He, Hanghang Tong, Heng Ji, Huan Zhang.
  [Arxiv]

• 

  [📢 oral] ShortageSim: Simulating Drug Shortages under Information Asymmetry

  Mingxuan Cui*, Yilan Jiang*, Duo Zhou*, Cheng Qian, Yuji Zhang†, Qiong Wang†.
  [AAAI 2026]

• 

  ModelingAgent: Bridging LLMs and Mathematical Modeling for Real-World Challenges

  Cheng Qian, Hongyi Du, Hongru Wang, Xiusi Chen, Yuji Zhang, Avirup Sil, Chengxiang Zhai, Kathleen McKeown, Heng Ji.
  [EMNLP 2025]

• 

  Internal Activation as the Polar Star for Steering Unsafe LLM Behavior

  Peixuan Han, Cheng Qian, Xiusi Chen, Yuji Zhang, Denghui Zhang, Heng Ji.
  [EMNLP 2025]

• 

  EscapeBench: Towards Advancing Creative Intelligence of Language Model Agents

  Cheng Qian, Peixuan Han, Qinyu Luo, Bingxiang He, Xiusi Chen, Yuji Zhang, Hongyi Du, Jiarui Yao, Xiaocheng Yang, Denghui Zhang, Yunzhu Li, Heng Ji.
  [ACL 2025]

• 

  Integrative Decoding: Improve Factuality via Implicit Self-consistency

  Yi Cheng, Xiao Liang, Yeyun Gong, Wen Xiao, Song Wang, Yuji Zhang, Wenjun Hou, Kaishuai Xu, Wenge Liu, Wenjie Li, Jian Jiao, Qi Chen, Peng Cheng, Wayne Xiong.
  [ICLR 2025]

• 

  EVEDIT: Event-based Knowledge Editing for Deterministic Knowledge Propagation

  Jiateng Liu*, Pengfei Yu*, Yuji Zhang, Sha Li, Zixuan Zhang, Heng Ji.
  [EMNLP 2024]

• 

  VIBE: Topic-Driven Temporal Adaptation for Twitter Classification

  Yuji Zhang, Jing Li, Wenjie Li.
  [EMNLP 2023]

• 

  # HowYouTagTweets: Learning User Hashtagging Preferences via Personalized Topic Attention

  Yuji Zhang, Yubo Zhang, Chunpu Xu, Jing Li, Ziyan Jiang, Baolin Peng.
  [EMNLP 2021]

Honors

• Outstanding Student Leader of Southeast University 2017, 2018, 2019
• Second Prize in National Undergraduate Mathematical Contest in Modeling, 2018
• Skyworth Scholarship, 2018
• National Scholarship (Top 1%, the highest scholarship for undergraduates), 2017
• Merit Student of Jiangsu Province (Top 0.1%), 2016

Experiences

• 2019 - 2020: Research intern at Dartmouth College (Advisors: Prof. Xingdong Yang, Prof. Te-Yen Wu)
• 2023 - 2024: Visiting PhD student at UIUC (Advisor: Prof. Heng Ji)

Professional Activities

Area Chair

• ACL 2025, LREC 2026

Conference Reviewer

• ICLR 2026, ACL 2025, EMNLP 2025, NAACL 2024, KDD 2023, COLING 2023

Instructor

• UIUC CS 591: Biologically Plausible Artificial Intelligence, Spring 2026, UIUC
• UIUC CS 591: Biologically Plausible Artificial Intelligence, Fall 2025, UIUC
• UIUC CS 591: Biologically Plausible Artificial Intelligence, Spring 2025, UIUC

Teaching Assistant

• COMP 3134: Business Intelligence and CRM, Fall 2023, PolyU
• COMP 5511: Artificial Intelligence Concepts, Spring 2023, PolyU
• COMP 5511: Artificial Intelligence Concepts, Fall 2022, PolyU
• COMP 4141: Crowdsourcing and e-Finance, Spring 2022, PolyU
• COMP 1433: Introduction to Data Analytics, Fall 2022, PolyU
• COMP 1433: Introduction to Data Analytics, Spring 2021, PolyU

Judge/Mentor

• Woman in Computer Science 2025 Flagship Hackathon, Code Ada: Code to the Future

Misc.

• I enjoy reading and crafts. I have practiced taekwondo and piano for four years.

• In my free time, I love hiking, swimming, and exploring various cultural destinations.

×