Wenqiang Zhang

OneTrackerV2: Unified Multimodal Visual Object Tracking with Mixture of Experts

LingoLoop Attack: Trapping MLLMs via Linguistic Context and State Entrapment into Endless Loops

Fast Visuomotor Policy for Robotic Manipulation

SGM: A Statistical Gödel Machine for Risk-Controlled Self-Improvement

Dynamic Semantic-Aware Correlation Modeling for UAV Tracking

Boosting Adversarial Transferability with Spatial Adversarial Alignment

ForceVLA: Enhancing VLA Models with a Force-aware MoE for Contact-rich Manipulation

Enhancing Diffusion-based Unrestricted Adversarial Attacks via Adversary Preferences Alignment

Agentic RL Scaling Law: Spontaneous Code Execution for Mathematical Problem Solving

General Compression Framework for Efficient Transformer Object Tracking

Empowering Teachers with Enhanced Knowledge via Variable Scale Distillation Framework

Generalizable Dynamic Radiance Field in Egocentric View

Image Generation with Channel-wise Quantization