Xiaomeng Hu

Xiaomeng Hu, Ph.D. student
Department of Computer Science and Engineering
The Chinese University of Hong Kong
Supervisor: Professor Tsung-Yi Ho

BIOGRAPHY

I am currently a second-year Ph.D. student at the Department of Computer Science and Engineering of The Chinese University of Hong Kong under the supervision of Prof. Tsung-Yi Ho and co-supervised by Dr. Pin-Yu Chen from IBM-Research.

My primary research interests are interpreting and controlling the behavior of LLMs through mechanism understanding and decoding intervention, as well as enhancing their complex reasoning capabilities for real-world applications like end-to-end agents and generative reward modeling.

I have extensive collaborations with industry researchers. Previously, I was mentored by Dr. Chenyan Xiong from Microsoft Research Redmond. Now I work closely with Dr. Pin-Yu Chen from IBM Research and Dr. Fei Huang from Alibaba Qwen Team.

NEWS

Jan. 2024 Accept ACL ARR 2025 February reviewer invitations.
Jan. 2024 Accept IJCAI 2025 reviewer invitations.
Dec. 2024 Token Highlighter was accepted by AAAI 2025 with an oral presentation (top 4.65%)
Sep. 2024 Accept ICLR 2025 reviewer invitations.
Sep. 2024 Gradient Cuff was accepted by NeurIPS 2024.
Apr. 2024 IBM Blog highlighted Gradient Cuff: What is red teaming for generative AI?
Sep. 2023 RADAR was accepted by NeurIPS 2023.
Jul. 2023 IBM Blog highlighted RADAR: What is AI forensics?
Apr. 2022 P3 Ranker was accepted by SIGIR 2022.

CONTACT

Email (Personal): greghxm [at] foxmail.com
Email (Campus): greghxm [at] link.cuhk.edu.hk

RESEARCH FOCUS (RECENT)

Large Language Models
Reinforcement Learning
Test-time Scaling
AI Safety

EDUCATION

Ph.D. Computer Science and Engineering, The Chinese University of Hong Kong, Aug. 2023 -
B.Eng. Artificial Intelligence, Northeastern University, Sep. 2019 - Jul. 2023

PUBLICATION

* denotes equal contribution.

Conference papers

[C4] Xiaomeng Hu, Pin-Yu Chen, Tsung-Yi Ho. Token Highlighter: Inspecting and Mitigating Jailbreak Prompts for Large Language Models. AAAI 2025 (Oral). (paper) (demo)
[C3] Xiaomeng Hu, Pin-Yu Chen, Tsung-Yi Ho. Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapes. NeurIPS 2024. (paper) (demo) (code) (IBM Blog)
[C2] Xiaomeng Hu, Pin-Yu Chen, Tsung-Yi Ho. RADAR: Robust AI-Text Detection via Adversarial Learning. NeurIPS 2023. (paper) (demo) (code) (IBM Blog)
[C1] Xiaomeng Hu*, Shi Yu*, Chenyan Xiong, Zhenghao Liu, Zhiyuan Liu, Ge Yu. P³ Ranker: Mitigating the Gaps between Pre-training and Ranking Fine-tuning with Prompt-based Learning and Pre-finetuning. SIGIR 2022. (paper) (code)

EXPERIENCE

International Business Machine, New York, USA, Feb. 2023 - Sept. 2023
Research Intern (Remote), IBM Research AI
Topic: AI-Text detection & Reinforcement Learning
Project: RADAR (NeurIPS 2023)
Tsinghua University, Beijing, P.R. China, Aug. 2021 - Jun. 2022
Research Intern, THU-NLP Lab
Topic: Information Retrieval and Pretrained Language Models
Project: P3 Ranker (SIGIR 2022)