SOURCE-LINKED INTELLIGENCE

Hyper3-CLIP: Hierarchy-Conditioned Hyperbolic Vision-Language Training

arXiv · AI, language, vision and robotics · article · Aug 29, 2026 · UTC

CLIP-like vision-language models (VLMs) trained with contrastive objectives learn strong global image-text representations, but their Euclidean embeddings and global pooling fail to encode relational structure such as part-whole and parent-child relations. Hyperbolic VLMs address this gap with entailment-based objectives, and text-conditioned variants improve fine-grained alignment through sentence- and phrase-level queries. However, these two lines of work remain separate: hyperbolic VLMs use static image and region features, while query-conditioned methods lack hierarchical geometric structu

Read original source ↗ Open in workspace

recordType

paper

region

Global