4 papers · 1 filter
Unlocking Fine-Grained Perception in CLIP via Structurally-Aware Latent Masked Modeling
Juntong Li, Lingwei Dang, Haomin Wu +3
Vision-Language Models (VLMs) such as CLIP excel in global semantic alignment but often lack fine-grained perceptual capabilities. This hinders dense prediction tasks and bottlenec…
StyleForge: Indoor Furniture Styling by Counterfactual Reasoning in a Hypergraph Field
Lingwei Dang, Shishuo Shang, Pan Liu +9
Fixed-layout indoor furniture styling requires selecting assets that form a coherent room without changing the prescribed furniture categories, positions, orientations, or scales.…
Not Blind but Silenced: Rebalancing Vision and Language via Adversarial Counter-Commonsense Equilibrium
Qingxin Xiao, Peilin Zhao, Yangyang Zhao +2
During MLLM decoding, attention often abnormally concentrates on irrelevant image tokens. While existing research dismisses this as invalid noise and forcibly redirects attention t…
Video Anomaly Detection with Semantics-Aware Information Bottleneck
Juntong Li, Lingwei Dang, Qingxin Xiao +5
Semi-supervised video anomaly detection methods face two critical challenges: (1) Strong generalization blurs the boundary between normal and abnormal patterns. Although existing a…