Zongqing Lu

DemoGrasp: Universal Dexterous Grasping from a Single Demonstration

Guided Policy Optimization under Partial Observability

Vision-Language-Action Pretraining from Large-Scale Human Videos

Guided Actor-Critic: Off-Policy Partially Observable Reinforcement Learning with Privileged Information

Multi-Agent Guided Policy Optimization

OpenT2M: No-frill Motion Generation with Open-source, Large-scale, High-quality Data

Predictive Embedding as Latent Action: Towards VLA Pretraining in the Wild

GateFlow: Mitigating Shortcut Learning in VLA Models via Gated Flow Matching

Leveraging Pre-Trained Tacit Model for Efficient Multi-Agent Coordination

RL from Physical Feedback: Aligning Large Motion Models with Humanoid Control

Temporal Difference Learning with Constrained Initial Representations

EgoPlan: Towards Effective Embodied Agents via Egocentric Planning

From Experts to a Generalist: Toward General Whole-Body Control for Humanoid Robots

OpenMMEgo: Enhancing Egocentric Understanding for LMMs with Open Weights and Data

Learning Video-Conditioned Policy on Unlabelled Data with Joint Embedding Predictive Transformer

Discrete Latent Plans via Semantic Skill Abstractions

Efficient Residual Learning with Mixture-of-Experts for Universal Dexterous Grasping

Cross-Domain Offline Policy Adaptation with Optimal Transport and Dataset Constraint

Watch Less, Do More: Implicit Skill Discovery for Video-Conditioned Policy

MLLM as Retriever: Interactively Learning Multimodal Retrieval for Embodied Agents

Cradle: Empowering Foundation Agents towards General Computer Control

Planning with Quantized Opponent Models

MEgoHand: Multimodal Egocentric Hand-Object Interaction Motion Generation

Revisiting Cooperative Off-Policy Multi-Agent Reinforcement Learning

Quo Vadis, Motion Generation? From Large Language Models to Large Motion Models

From Pixels to Tokens: Byte-Pair Encoding on Quantized Visual Modalities

Scaling Large Motion Models with Million-Level Human Motions

Learning Diverse Bimanual Dexterous Manipulation Skills from Human Demonstrations

Cross-Embodiment Dexterous Grasping with Reinforcement Learning

SELU: Self-Learning Embodied MLLMs in Unknown Environments

Cradle: Empowering Foundation Agents towards General Computer Control

Best Possible Q-Learning

Egocentric Vision Language Planning

Offline Model-Based Skill Stitching

GAMEBOT: Gaming Arena for Model Evaluation - Battle of Tactics

$f$-Divergence Policy Optimization in Fully Decentralized Cooperative MARL