Yong Li

11 papers Tsinghua University

• AgentRecBench: Benchmarking LLM Agent-based Personalized Recommender Systems
• Balanced Token Pruning: Accelerating Vision Language Models Beyond Local Optimization
• Diffusion Transformers as Open-World Spatiotemporal Foundation Models
• LLM-Explorer: A Plug-in Reinforcement Learning Policy Exploration Enhancement Driven by Large Language Models
• PID-controlled Langevin Dynamics for Faster Sampling on Generative Models
• RoboScape: Physics-informed Embodied World Model
• Satellites Reveal Mobility: A Commuting Origin-destination Flow Generator for Global Cities
• Skrull: Towards Efficient Long Context Fine-tuning through Dynamic Data Scheduling
• Sparse Diffusion Autoencoder for Test-time Adapting Prediction of Complex Systems
• Synergistic Tensor and Pipeline Parallelism
• TrajAgent: An LLM-Agent Framework for Trajectory Modeling via Large-and-Small Model Collaboration