AI Paper Atlas

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

Variational Adapter for Cross-modal Similarity Representation

WenZhang Wei, Zhipeng Gui, Dehua Peng, Tiandi Ye, Huayi Wu

ICML 2026 · Vision-language and multimodal › Cross-modal learning · approach: Probabilistic embeddings for cross-modal retrieval · 0 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
use VACSR instead of PCME++CLIP ViT-B/32 · COCO 5K image-to-textR@1: 62.1 -> 66.5Improved
use VACSR instead of PCME++CLIP ViT-B/16 · ECCV Caption image-to-textmAP@R: 34.5 -> 35.4Improved
use VACSR instead of NPC at 20% noiseCLIP ViT-B/32 · COCO noisy correspondenceRSUM: 529.8 -> 539.0Improved
use VACSR under 50% noise versus next-bestCLIP ViT-B/32 · COCO noisy correspondenceRSUM: 523.4 -> 534.2Improved
use VACSR instead of MMRLCLIP ViT-B/16 · ImageNet OOD averageaccuracy: 60.6 -> 61.6Improved
use VACSR instead of CoPrompt on novel classesCLIP ViT-B/16 · Base-to-novel 11 datasetsnovel accuracy: 77.23 -> 76.08Hurt
use VACSR instead of MMRL on harmonic meanCLIP ViT-B/16 · Base-to-novel 11 datasetsharmonic mean: 81.20 -> 80.37Hurt
add Gaussian mixture (GMM) prior instead of single Gaussian (ablation)VACSR · ECCV CaptionmAP@R: 39.6 -> 40.7Improved
add distributional optimization loss L_sigma (ablation)VACSR · ECCV CaptionmAP@R: 39.8 -> 40.7Improved
add L_sigma (ablation)VACSR · CxCR@1: 60.7 -> 59.7Hurt
add KL regularization (ablation)VACSR · ECCV CaptionmAP@R: 40.1 -> 40.7Improved
add sigmoid normalization of variance (ablation)VACSR · ECCV CaptionmAP@R: 40.1 -> 40.7Improved

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