Huan-ang Gao

RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation

How Far Can Unsupervised RLVR Scale LLM Training?

CubeBench: Diagnosing Interactive, Long-Horizon Physical Intelligence under Partial Observations

Sim2Real-HOI: Sim-to-Real HOI Video Generation via Decoupled Motion–Appearance Diffusion

Straight-Line Diffusion Model for Efficient 3D Molecular Generation

Ctrl-U: Robust Conditional Image Generation via Uncertainty-aware Reward Modeling

Improving Zero-Shot Generalization of Instruction Tuning by Data Arrangement

SPA: Enhancing 3D Multimodal LLMs with Mask-based Streamlining Preference Alignment

RGM: Reconstructing High-fidelity 3D Car Assets with Relightable 3D-GS Generative Model from a Single Image