Chenxi Liu

I am a fourth-year Ph.D. student in Computer Science at the University of Maryland, College Park, advised by Prof. Heng Huang. I am currently a Research Intern at Meta FAIR, building agentic systems for real-time streaming video generation. Previously, I interned at ByteDance, working on reinforcement learning for LLM reasoning and multi-agent security.

My research interests include:

LLM/VLM Reasoning
Reinforcement Learning
Agentic Systems
World Models
Recursive Self-Improvement

News

Sep 2026Three papers accepted to NeurIPS 2026: EVAN (first author), MPQ, and AutoTTS.
May 2026Joined Meta FAIR as a Research Intern, working on agentic systems for real-time streaming video generation.
Apr 2026Multi-Crit is selected as a Highlight paper at CVPR 2026.
Feb 2026Multi-Crit is accepted to CVPR 2026.
Jan 2026New preprint: Learning from Self-Debate, preparing reasoning models for multi-agent debate.
Nov 2025New preprint: ERPO, exploring the residual prompts left behind by GRPO.

Earlier news

Jun 2025Joined ByteDance as a Research Intern, working on RL for LLM reasoning and multi-agent security.
Jun 2025Federated Continuous Category Discovery and Learning is accepted to ICCV 2025.
Jul 2024ASP for few-shot class-incremental learning is accepted to ECCV 2024.
May 2024Self-Filter is accepted to Findings of ACL 2024.
Aug 2023Started my Ph.D. in Computer Science at the University of Maryland, College Park.
Jan 2023Deja Vu is accepted to ICLR 2023.

Selected Publications

Grounding Multimodal Reasoning with Evidence-Ablated Negatives

Chenxi Liu, Tianyi Xiong, Yanshuo Chen, Heng Huang

NeurIPS 2026Paper Code

MPQ: A Message-Passing View of Post-Training Quantization

Yanshuo Chen, Chenxi Liu, Reza Shirkavand, Heng Huang

NeurIPS 2026Paper Code

LLMs Improving LLMs: Agentic Discovery for Test-Time Scaling

Tong Zheng, Haolin Liu, Chengsong Huang, Huiwen Bao, Sheng Zhang, Rui Liu, Runpeng Dai, Ruibo Chen, Chenxi Liu, Tianyi Xiong, Xidong Wu, Hongming Zhang, Heng Huang

NeurIPS 2026Paper Code

Learning from Self-Debate: Preparing Reasoning Models for Multi-Agent Debate

Chenxi Liu, Yanshuo Chen, Ruibo Chen, Tianyi Xiong, Tong Zheng, Heng Huang

arXiv 2026Paper Code

Black-Box Red-Teaming of Multi-Agent Systems via Reinforcement Learning

Yuqi Jia, Hangsheng Zhang, Chenxi Liu, Junjie Liang, Xun Chen

arXiv 2026Paper

Multi-Crit: Benchmarking Multimodal Judges on Pluralistic Criteria-Following

Tianyi Xiong*, Yi Ge*, Ming Li, Zuolong Zhang, Pranav Kulkarni, Kaishen Wang, Qi He, Zeying Zhu, Chenxi Liu, Ruibo Chen, Tong Zheng, Yanshuo Chen, Xiyao Wang, Renrui Zhang, Wenhu Chen, Heng Huang

CVPR 2026HighlightPaper Project

Explore Data Left Behind in Reinforcement Learning for Reasoning Language Models

Chenxi Liu, Junjie Liang, Yuqi Jia, Bochuan Cao, Yang Bai, Heng Huang, Xun Chen

arXiv 2025Paper Code

Federated Continuous Category Discovery and Learning

Lixu Wang*, Chenxi Liu*, Junfeng Guo, Qingqing Ye, Heng Huang, Haibo Hu, Wei Dong

ICCV 2025Paper

OPTune: Efficient Online Preference Tuning

Lichang Chen, Jiuhai Chen, Chenxi Liu, John Kirchenbauer, Davit Soselia, Chen Zhu, Tom Goldstein, Tianyi Zhou, Heng Huang

arXiv 2024Paper

Few-Shot Class Incremental Learning with Attention-Aware Self-Adaptive Prompt

Chenxi Liu, Zhenyi Wang, Tianyi Xiong, Ruibo Chen, Yihan Wu, Junfeng Guo, Heng Huang

ECCV 2024Paper Code

Your Vision-Language Model Itself Is a Strong Filter: Towards High-Quality Instruction Tuning with Data Selection

Ruibo Chen, Yihan Wu, Lichang Chen, Guodong Liu, Qi He, Tianyi Xiong, Chenxi Liu, Junfeng Guo, Heng Huang

ACL 2024 FindingsPaper Code

Deja Vu: Continual Model Generalization for Unseen Domains

Chenxi Liu*, Lixu Wang*, Lingjuan Lyu, Chen Sun, Xiao Wang, Qi Zhu

ICLR 2023Paper Code

* Equal contribution. Full list on Google Scholar.

Experience

Meta FAIR
Research Intern · May 2026 – Present
Agentic systems for streaming video generation, distilled into the video model with on-policy self-distillation for real-time generation.
ByteDance
Research Intern · Jun 2025 – Sep 2025
Reinforcement learning for LLM reasoning (ERPO) and multi-agent security (ReMAS).

Education

Ph.D. in Computer Science, University of Maryland, College Park
Aug 2023 – Present · Advisor: Prof. Heng Huang
M.S. in Electrical Engineering, Northwestern University
Aug 2021 – Mar 2023 · Advisors: Prof. Qi Zhu and Prof. Xiao Wang
B.S. in Opto-Electronics Information Science and Engineering, Sun Yat-sen University
Sep 2016 – Jun 2020 · Advisor: Prof. Xiao Zhang