Deqiang Jiang

VITA-E: A Dual-Model Framework for Real-Time, Interruptible, and Concurrent Human-Robot Interaction

RAR: Reversing Visual Attention Re-Sinking for Unlocking Potential in Multimodal Large Language Models

VITA-VLA: Efficiently Teaching Vision-Language Models to Act via Action Expert Distillation

TACO: Think-Answer Consistency for Optimized Long-Chain Reasoning and Efficient Data Learning via Reinforcement Learning in LVLMs

Dimensional Debiasing via Multi-Agent Correction

Non-targeted Adversarial Attacks on Vision-Language Models via Maximizing Information Entropy