CV

Download CV (PDF)

Education

• Ph.D. in Computer Science and Engineering, The Chinese University of Hong Kong, 2024-2028 (expected)
• M.S. in Computer Science, National University of Singapore, 2021-2024
• B.S. in Computer Science, Shanghai Jiao Tong University, 2017-2021 (Graduated with Honors)

Research Interests

• Multimodal Large Language Models
• Unified Vision-Language Modeling
• World Models and Action Models

Research Experience

Skywork AI - Research Intern May 2026 - Present

• Led the memory module design for Matrix Game 3.5, including dynamic object filtering and object-token conditioning for long-horizon scene and protagonist consistency.
• Developed a video-action Mixture-of-Transformers architecture with autoregressive causal masking for interactive policy execution.
• Leading Matrix Game Next, a neural rendering system for real-time game visual enhancement while preserving gameplay-relevant structure and motion.

Tencent - Research Intern July 2025 - January 2026

• Developed a controllable game video generation framework that separates semantic layout prediction from pixel-level rendering.
• Designed a global memory bank to maintain persistent world state and long-term spatiotemporal consistency.

Microsoft - Research Intern October 2024 - June 2025

• Studied cross-task generalization between understanding and generation in unified VLMs, identifying visual-space alignment as a key factor in knowledge transfer.

Publications

Less Is More: Vision Representation Compression for Efficient Video Generation with Large Language Models

Cross-Task Generalization Between Understanding and Generation in Unified Vision-Language Models: A Controlled Study

CLIP-MoE: Towards Building Mixture of Experts for CLIP with Diversified Multiplet Upcycling

SURf: Teaching Large Vision-Language Models to Selectively Utilize Retrieved Information

Learning the Unlearned: Mitigating Feature Suppression in Contrastive Learning

Birds of a feather trust together: Knowing when to trust a classifier via adaptive neighborhood aggregation

EatingTrak: Detecting Fine-grained Eating Moments in the Wild Using a Wrist-mounted IMU

Learning generative RNN-ODE for collaborative time-series and event sequence forecasting

Synergetic learning of heterogeneous temporal sequences for multi-horizon probabilistic forecasting

Academic Service

Conference refereeing:

• NeurIPS 2024
• NeurIPS 2025
• ARR May 2025
• ICLR 2026