Chuyue Li
Computer Science Undergraduate
Multimodal AI Agents
Multimodal Generative Models
GenAI
Agentic RL
LLM/VLMs
CV
ML
RL
🎓 Applying for Fall 2026 PhD / Research Master's Programs
📧 lichuyue0312 [at] gmail [dot] com
📞 (+86)181-1636-9770
🌐 Homepage
📍 Berkeley, CA / Shanghai, China
👋 About Me
Hey there! 👋 I'm Chuyue, a senior CS student at ShanghaiTech University, currently conducting research at Tencent Hunyuan.
My research interests focus on Multimodal AI agents, Multimodal Generative Models, GenAI, and Agentic RL, along with related areas in machine learning and computer vision. I'm passionate about exploring the intersection of these fields and developing innovative solutions that bridge the gap between different modalities and intelligent agent systems.
I am currently applying for Fall 2026 PhD or Research Master's programs. If you have any suitable opportunities or potential collaborations, I would greatly appreciate your consideration and would love to hear from you! Thank you very much for your time and interest!
My research interests focus on Multimodal AI agents, Multimodal Generative Models, GenAI, and Agentic RL, along with related areas in machine learning and computer vision. I'm passionate about exploring the intersection of these fields and developing innovative solutions that bridge the gap between different modalities and intelligent agent systems.
I am currently applying for Fall 2026 PhD or Research Master's programs. If you have any suitable opportunities or potential collaborations, I would greatly appreciate your consideration and would love to hear from you! Thank you very much for your time and interest!
🔬 Research Interests
Multimodal AI Agents
Multimodal Generative Models
Agentic RL
LLM/VLM
Generative AI
Machine Learning
Reinforcement Learning
Computer Vision
🎓 Education
University of California, Berkeley (UCB)
CS Exchange Student • GPA: 3.9/4.0
September 2024 - May 2025
ShanghaiTech University
B.S. in Computer Science
September 2022 - Present
📝 Publications
ICML 2025
Innovatively designed a framework to enable finer-grained cross-modal semantic alignment and control in generating time series data from unstructured text.
Proposed GENTS: unifying time series conditional generation, forecasting, and editing into a masked-conditional generation problem within a unified framework.
💼 Professional Experience
Applied Research Intern
Tencent Hunyuan - Large Multimodal Model Department, Visual Agent Team
June 2025 - Present
Working on multi-level auto-planning interactive visual generative agent systems with focus on evaluation metrics and reward functions design.
Research Assistant
ShanghaiTech VDI Center
June 2024 - Present
Conducting research on multimodal time series generation and analysis using diffusion models and large language models.
Research Assistant
CircleCat
February 2025 - July 2025
Working on LLM-based code error classification under supervision of Google DeepMind and Duolingo researchers.
Software Engineer Intern
CircleCat
November 2024 - March 2025
Developed and optimized software systems with focus on large language model applications and code analysis tools.
🚀 Projects
Large Model Stepwise Text-to-3D Generation Optimization
Best Popularity Award
NYUSHDIC
- Lead the design and implementation of a multi-turn interactive, dimensionality-elevating agent system for generating 3D visual deliverables from textual input.
- Construct a 2D bridge based on text-to-image transformers to mitigate large modality gaps in diffusion-based text-to-3D generation.
- Utilize CoT and multi-agent collaboration framework supporting multi-turn dialogue and iterative modification based on user feedback.