Skip to main content
Search

# Cross-Image Comparison

1 related articles

Subscribe to updates
  1. arXiv

    Vision-language models see the match but cannot say it: training 16 to 32 cross-image heads lifts SPair-71k from 0.430 to 0.712

    Vision-language models often fail to match a point or object across two images even when the answer is decodable from their hidden states. Across Qwen3-VL, InternVL3.5 and Gemma-3, the failure splits in two: exact copies are compared but not read out at the answer position, while semantic correspondence sits in the visual tokens and is never compared. Training only the query and key projections of 16 to 32 cross-image attention heads raises Qwen3-VL-8B on SPair-71k from 0.430 to 0.712 and transfers to HPatches and BLINK, which were not used for training.

You’re all caught up