Yaowei Wang

UI-MOPD achieves a remarkable balance between retaining existing capabilities and adapting to new platforms, with task success rates that challenge conventional approaches in GUI agent learning.

Unreliable pseudo-labels in object detection can be transformed into reliable training signals, leading to substantial performance gains across domains.

ICMPG achieves a groundbreaking balance between semantic fidelity and physical realism in motion synthesis, outperforming traditional methods in both standard and zero-shot scenarios.

MemVenom reveals that web agents can be compromised with up to 99.15% success through sophisticated memory poisoning attacks that bypass traditional defenses.

Removing objects from video just got a whole lot cleaner: GenEraser doesn't just erase the object, it intelligently removes associated effects like shadows and reflections, setting a new bar for realistic video editing.

Unlock "white-box" reasoning in vision-language models: SegCompass's sparse autoencoder creates an interpretable bridge between visual perception and chain-of-thought, outperforming black-box alignment methods.

Steer LVLMs' attention with caption guidance and watch object hallucinations drop by 6%—no training required.

Adversarial training of large vision models doesn't have to break the bank: CAAT achieves comparable robustness to standard methods by tuning just 6% of the parameters.