AI Paper Atlas

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

Spherical Linear Interpolation and Text-Anchoring for Zero-shot Composed Image Retrieval

Young Kyun Jang, Dat B Huynh, Ashish Shah, Wen-Kai Chen, Ser-Nam Lim

ECCV 2024 · Vision-language and multimodal › CLIP and zero-shot · approach: Attribute-based zero-shot learning · 45 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 TAT (text-anchored LoRA tuning of image encoder) before SlerpSlerp ZS-CIR · CIRCO test, CLIP-ViT-L/14mAP@5: 8.76 -> 18.46Improved
add TAT before SlerpSlerp ZS-CIR · CIRR test, CLIP-ViT-L/14Recall@1: 24.43 -> 30.94Improved
fix text encoder and tune image encoder (text-anchoring) rather than tune both encoders (ablation)TAT · CIRCO test, CLIP-ViT-L/14mAP@5: 8.26 -> 18.46Improved
fix text encoder and tune image encoder rather than image-anchoring (ablation)TAT · CIRCO test, CLIP-ViT-L/14mAP@5: 7.54 -> 18.46Improved
scale TAT training data from Laion-2M to Laion-4M or 8M (ablation)TAT · CIRCO test, CLIP-ViT-L/14mAP@5: 18.46 -> 18.45 (4M), 18.23 (8M)No clear effect
train TAT on only 0.58M LLaVA-Align pairs (ablation)TAT · CIRCO test, CLIP-ViT-L/14mAP@5: 17.05 vs 18.46 with Laion-2MNo clear effect
train Pic2Word on Laion-2M instead of CC3M (ablation)Pic2Word · CIRCO test, CLIP-ViT-L/14mAP@5: 8.93 vs Pic2Word 8.72No clear effect
apply LoRA to Pic2Word (ablation)Pic2Word · CIRCO test, CLIP-ViT-L/14mAP@5: 8.89 vs Pic2Word 8.72No clear effect
raise Slerp text weight alpha up to 0.8 then to 1.0 (no image) (ablation)Slerp · CIRCO test, CLIP-ViT-L/14mAP and RecallMixed
replace original CLIP with TAT-trained CLIP in supervised Combiner trainingCombiner · CIRR test, CLIP-ViT-L/14Recall@1: 30.31 -> 36.58Improved
replace original CLIP with TAT-trained CLIP in supervised Combiner trainingCombiner · FashionIQ val, CLIP-ViT-L/14Average R@10: 35.64 -> 36.69Improved
add TAT on top of SlerpSlerp ZS-CIR · FashionIQ val, CLIP-ViT-B/32Average R@10: 23.75 -> 22.96Mixed

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