7 papers · 1 filter
Can Cross-Layer Transcoders Replace Vision Transformer Activations? An Interpretable Perspective on Vision
Gerasimos Chatzoudis, Konstantinos D. Polyzos, Zhuowei Li +4
Understanding the internal activations of Vision Transformers (ViTs) is critical for building interpretable and trustworthy models. While Sparse Autoencoders (SAEs) have been used…
Decoupling Vision and Language: Codebook Anchored Visual Adaptation
Jason Wu, Tianchen Zhao, Chang Liu +7
Large Vision-Language Models (LVLMs) use their vision encoders to translate images into representations for downstream reasoning, but the encoders often underperform in domain-spec…
Beyond Interpretability: When, Why, and How Sparse Autoencoders Enable Label-Free Visual Steering
Gerasimos Chatzoudis, Zhuowei Li, Gemma E. Moran +2
Sparse Autoencoders (SAEs) are increasingly used to interpret foundation models, but their role as an actionable intervention space remains less understood, especially in vision. W…
Optimal Transport-Guided Source-Free Adaptation for Face Anti-Spoofing
Zhuowei Li, Tianchen Zhao, Xiang Xu +7
Developing a face anti-spoofing model that meets the security requirements of clients worldwide is challenging due to the domain gap between training datasets and diverse end-user…
Show and Segment: Universal Medical Image Segmentation via In-Context Learning
Yunhe Gao, Di Liu, Zhuowei Li +4
Medical image segmentation remains challenging due to the vast diversity of anatomical structures, imaging modalities, and segmentation tasks. While deep learning has made signific…
The Hidden Life of Tokens: Reducing Hallucination of Large Vision-Language Models via Visual Information Steering
Zhuowei Li, Haizhou Shi, Yunhe Gao +7
Large Vision-Language Models (LVLMs) can reason effectively over both textual and visual inputs, but they tend to hallucinate syntactically coherent yet visually ungrounded content…
MLLM-as-a-Judge for Image Safety without Human Labeling
Zhenting Wang, Shuming Hu, Shiyu Zhao +12
Image content safety has become a significant challenge with the rise of visual media on online platforms. Meanwhile, in the age of AI-generated content (AIGC), many image generati…