AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Knowledge Transfer from Interaction Learning

Yilin Gao, Kangyi Chen, Zhongxing Peng, Hengjie Lu, Shugong Xu

ICCV 2025 · Vision-language and multimodal › CLIP and zero-shot · approach: Open-vocabulary alignment from VLMs · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add interaction queries supervised by VLM interactions with gated control to ViT-LViT-L/16 · TinyImageNettop-1 accuracy: 87.83 -> 91.08Improved
add interaction queries to the supervised I-ViT (ablation)I-ViT-L/16 · TinyImageNettop-1 accuracy: 80.7 -> 91.1Improved
add gated control network to fuse interactions (ablation)I-ViT-L/16 · TinyImageNettop-1 accuracy: 90.3 -> 91.1Improved
add VLM interaction constraint (supervision of C_VLM) on interaction queries (ablation)I-ViT-L/16 · TinyImageNettop-1 accuracy: 89.6 -> 91.1Improved
supervise original queries directly with VLM interactions (ablation)ViT-L/16 · TinyImageNettop-1 accuracy: 87.8 -> 80.7Hurt
add VLM interactions on limited training dataViT-L/16 from scratch · TinyImageNet 10% datatop-1 accuracy: 12.64 -> 15.60Improved
add VLM interactions to DETRDETR · COCO 2017 detectionmAP: 42.0 -> 43.6Improved
add VLM interactions to DETR segmentationDETR · COCO 2017 segmentationAP mask: 31.1 -> 33.5Improved
add visual bounding-box prompts when extracting VLM interactions (ablation)VLM interaction extraction · COCO val2017mAP: 41.6 -> 43.6Improved
extract VLM interaction with foreground-background prompt instead of description prompt (ablation)I-ViT-S/16 · TinyImageNettop-1 accuracy: 80.59 -> 81.52Improved
add VLM interactions on sketch domainViT-L/16 · PACS Sketch zero-shotaccuracy: 15.49 -> 15.18No clear effect
add VLM interactions to small object detectionConditional-DETR · VOC few-shotAP_S: 5.0 -> 4.9No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy