Dongfang Liu

9

Papers

132

Total Citations

Papers (9)

Text Is MASS: Modeling as Stochastic Embedding for Text-Video Retrieval

Visual Agents as Fast and Slow Thinkers

AMD: Automatic Multi-step Distillation of Large-scale Vision Models

ProMotion: Prototypes As Motion Learners

Robust Ego-Exo Correspondence with Long-Term Memory

NeurIPS 2025arXiv

All You Need is One: Capsule Prompt Tuning with a Single Vector

Prototypical Transformer As Unified Motion Learners

BadPart: Unified Black-box Adversarial Patch Attacks against Pixel-wise Regression Tasks

Exploring the Adversarial Vulnerabilities of Vision-Language-Action Models in Robotics