Hang Su

DiffusionNFT: Online Diffusion Reinforcement with Forward Process

Exploratory Diffusion Model for Unsupervised Reinforcement Learning

Theoretical Analysis of Relative Errors in Gradient Computations for Adversarial Attacks with CE Loss

Helix: Evolutionary Reinforcement Learning for Open-Ended Scientific Problem Solving

Exploring the Secondary Risks of Large Language Models

AnyPos: Automated Task-Agnostic Actions for Bimanual Manipulation

Towards Safe Reasoning in Large Reasoning Models via Corrective Intervention

MePo: Meta Post-Refinement for Rehearsal-Free General Continual Learning

Vidarc: Low Latency Embodied Video Diffusion Model with Closed-loop Control

Vidar: Embodied Video Diffusion Model for Generalist Manipulation

Toward Guidance-Free AR Visual Generation via Condition Contrastive Alignment

RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation

Graph Diffusion for Robust Multi-Agent Coordination

Visual Generation Without Guidance

ManiBox: Enhancing Spatial Grasping Generalization via Scalable Simulation Data Generation

Accelerating PDE-Constrained Optimization by the Derivative of Neural Operators

Towards Mitigating Factual Hallucination in LLMs through Self-Alignment with Memory

CAPGen: An Environment-Adaptive Generator of Adversarial Patches