About Me
I am a Ph.D. student in Computer Science and Technology at the State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences (ICT, CAS), advised by Prof. Yuanzhuo Wang, Prof. Bingbing Xu, and Prof. Huawei Shen. I received my bachelor’s degree in Computer Science and Technology from China Agricultural University in 2023.
My research focuses on multimodal understanding, LLM post-training, multimodal generation, and agents. I am particularly interested in how models understand continuous audio-visual streams, learn from richer feedback, and interact with people in real time.
I am currently a research intern in the Visual Technology Department, Qwen Business Group at Alibaba, where I work with Jiaming Liu on proactive streaming video understanding and reinforcement learning for creative generation. Previously, I worked on robot interaction, reasoning, and planning at ByteDance Seed-Robotics.
- Multimodal Understanding
- LLM Post-training
- Multimodal Generation
- Agents
News
- May 2026 — Three papers accepted to ACL 2026: Learning from Mistakes (Main), ROMA (Findings), and Chain-of-Memory (Main, Oral).
- Apr. 2026 — Joined Alibaba’s Qwen Business Group as a multimodal research intern.
- July 2025 — MIGE accepted to ACM Multimedia 2025.
- Sept. 2024 — Joined ByteDance Seed-Robotics as a research intern.
Research & Publications
Selected First-Author Papers
Learning from Mistakes: Negative Reasoning Samples Enhance Out-of-Domain Generalization
Negative reasoning samples improve out-of-domain generalization; GLOW adaptively reweights training examples using their learning dynamics.
ROMA: Real-time Omni-Multimodal Assistant with Interactive Streaming Understanding
A real-time omni-multimodal assistant that supports proactive alerts, live narration, and reactive question answering over audio-video streams.
MIGE: Mutually Enhanced Multimodal Instruction-Based Image Generation and Editing
A unified framework for subject-driven generation and instruction-based image editing, with transfer to compositional editing tasks.
SyncPro: Synchronous Reasoning and Responding for Proactive Streaming Video Understanding
Proactive streaming video understanding with synchronized thinking and response streams, trained with supervised fine-tuning and reinforcement learning.
RARE: Reinforcement Learning with Adaptive Rubric Evolution for Open-Ended Generation
Adaptive rubric evolution for open-ended generation, using current policy feedback to replace saturated criteria while retaining other criteria and scores.
Collaborative Work
AVA-Encoder: Towards Agent-Native Video Representation Learning
Agent-native video representations that connect film understanding, structured knowledge, and video reconstruction.
DiffImaginE: Imagine to Verify Entity Types with Diffusion
Type-conditioned diffusion for verifying entity types using multimodal evidence.
Chain-of-Memory: Lightweight Memory Construction with Dynamic Evolution for LLM Agents
Lightweight memory construction and dynamic memory-chain evolution for long-term reasoning in LLM agents.
Beyond Entangled Planning: Task-Decoupled Planning for Long-Horizon Agents
Task-decoupled planning with global supervision for long-horizon agents.
Towards Robust Process Reward Modeling via Noise-aware Learning
Noise-aware learning and reflection-aware label correction for more robust process reward models.
GIFT: Games as Informal Training for Generalizable LLMs
Game-based informal learning and coordinated subtask training to improve LLM generalization.
Robix: A Unified Model for Robot Interaction, Reasoning and Planning
A unified high-level robot model for language interaction, reasoning, and long-horizon task planning.
Earlier publication
An Optimization Method for the Layout of Soil Humidity Sensors Based on Compressed Sensing
Experience
Alibaba · Qwen Business Group
Multimodal Research Intern, Visual Technology Department
Apr. 2026 – Present · Work with Jiaming Liu
- SyncPro: Proactive streaming video understanding with separate, synchronized thinking and response streams; trajectory construction, supervised fine-tuning, reinforcement learning, and evaluation.
- RARE: Reinforcement learning with adaptive rubric evolution for open-ended generation; reward design, model training, and evaluation on script generation and broader writing tasks.
ByteDance · Seed-Robotics
Research Intern, Multimodal LLMs for Robotics
Sept. 2024 – Apr. 2026
- Contributed to Robix, a unified model for robot interaction, reasoning, and planning.
- Curated approximately 130K robot task-planning examples and helped design and collect long-horizon mobile manipulation tasks in kitchen, supermarket, and tabletop settings.
- Developed consistency rewards and reward filtering for GRPO training to improve alignment between planning and execution.
Earlier experience
- Peking University, Research Intern, Jan. – Apr. 2022. Ubiquitous computing with Wi-Fi signals, advised by Prof. Daqing Zhang and Dr. Yang Li.
- China Agricultural University, Undergraduate Research Team Lead, Dec. 2019 – Mar. 2021. Ubiquitous computing for VR motion sickness.
Education
- Institute of Computing Technology, Chinese Academy of Sciences
Ph.D. in Computer Science and Technology · Sept. 2023 – June 2028 (expected) - China Agricultural University
Bachelor’s degree in Computer Science and Technology · Sept. 2019 – June 2023
Honors & Awards
- National Scholarship for Graduate Students (研究生国家奖学金) · 2026
- 领航博士生奖, ICT, CAS · 2025–2026 (1/45)
- Outstanding Merit Student (三好学生标兵), University of Chinese Academy of Sciences
- Outstanding Winner, Mathematical Contest in Modeling (MCM/ICM) · 2021
- First Prize, Beijing Region, China Undergraduate Mathematical Contest in Modeling · 2020
- Outstanding Student Scholarship, Second-Class Academic Excellence Scholarship, and Merit Student, China Agricultural University