Conference

We propose a framework for adaptive humanoid control via multi-behavior distillation and reinforced fine-tuning, achieving state-of-the-art performance and AAAI-26 Oral recommendation.

We present VMS, a unified whole-body controller that enables humanoid robots to learn diverse and dynamic behaviors within a single policy through hybrid tracking and orthogonal mixture of experts.

We propose Align-Then-stEer (ATE), a framework that adapts VLAs to novel robots and tasks through unified latent guidance. ATE can handle significant domain shifts without compromising performance and compatible to Pi0, RDT, and etc.

We introduce X-Loco, a framework for training a vision-based generalist humanoid locomotion controller through synergetic policy distillation.

We address humanoid skateboarding, a highly challenging task requiring stable dynamic maneuvering on an humanoid platform via physics-aware whole-body control for humanoid skateboarding in dynamic settings.

V2A unifies dynamics alignment, value alignment, and value assignment for heterogeneous cross-domain offline RL and improves robust source-data filtering.

HALO introduces a MuJoCo XLA-based two-stage identification pipeline that closes the heavy-load sim-to-real gap for agile humanoid skills.

Re^2MoGen combines LLM reasoning, keyframe-guided completion, and physics-aware RL refinement for open-vocabulary text-to-motion generation.

KineBench introduces the first IDM-free closed-loop benchmark for embodied world models, using explicit 3D kinematic grounding and robot-centric metrics for physical validation.

We introduce PRTS, a New Generation of Reinforcement Learning-Native Robotic Vision-Language-Action(VLA) foundation model.