AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Causal Disentanglement and Cross-Modal Alignment for Enhanced Few-Shot Learning

Tianjiao Jiang, Zhen Zhang, Yuhang Liu, Javen Qinfeng Shi

ICCV 2025 · Vision-language and multimodal › Cross-modal learning · approach: Unimodal-to-multimodal representation transfer · 4 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 ICA disentanglement of cache keys (ablation)CCA-FT · 11-dataset average, 1-shotaccuracy: 64.94 -> 66.00Improved
add ICA disentanglement of cache keys (ablation)CCA-FT · 11-dataset average, 16-shotaccuracy: 75.95 -> 77.60Improved
fine-tune the cache adapter (ablation)CCA-FT · 11-dataset average, 16-shotaccuracy: 75.75 -> 77.60Improved
fine-tune the CLIP text classifier (ablation)CCA-FT · 11-dataset average, 16-shotaccuracy: 77.00 -> 77.60Improved
add cross-attention fused features (ablation)CCA-FT · 11-dataset average, 16-shotaccuracy: 77.43 -> 77.60Improved
add cross-attention fused features (ablation)CCA-FT · 11-dataset average, 1-shotaccuracy: 65.81 -> 66.00Improved
fine-tune CCA (CCA-FT) instead of training-free CCA (ablation)CCA · ImageNet 16-shotaccuracy: 63.00 -> 66.04Improved
replace Tip-Adapter by training-free CCATip-Adapter · ImageNet-Sketchaccuracy: 35.71 -> 36.60Improved
replace Tip-Adapter-F by CCA-FT under Gaussian noiseTip-Adapter-F · Gaussian noise, 16-shot averageaccuracy: 60.04 -> 62.48Improved
replace Tip-Adapter-F by CCA-FT under FGSM attackTip-Adapter-F · FGSM, 16-shot averageaccuracy: 33.21 -> 35.48Improved
replace Tip-Adapter-F by CCA-FT on ImageNet-SketchTip-Adapter-F · ImageNet-Sketchaccuracy: 35.00 -> 35.87Improved
train CCA-FT 20 epochs instead of CoOp 200 epochsCoOp · ImageNet 16-shotaccuracy: 7.5h -> 4.9min; 62.95 -> 66.04Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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