Nanning Zheng

MRVF: Multi-Round Value Factorization with Guaranteed Iterative Improvement for Multi-Agent Reinforcement Learning

Decoupling Primitive with Experts: Dynamic Feature Alignment for Compositional Zero-Shot Learning

ChangingGrounding: 3D Visual Grounding in Changing Scenes

Vision Foundation Models Can Be Good Tokenizers for Latent Diffusion Models

Beyond Policy Training: Solution Search via Task Flow Learning and Planning from Unlabeled Videos

Information-Preserving Reformulation of Reasoning Traces for Antidistillation

Riemannian Consistency Model

VideoVLA: Video Generators Can Be Generalizable Robot Manipulators

On the Statistical Mechanisms of Distributional Compositional Generalization

HRVMamba: High-Resolution Visual State Space Model for Dense Prediction

Common 7B Language Models Already Possess Strong Math Capabilities