AI Paper Atlas

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

Rethinking Cross-Modal Anchor Alignment for Mitigating Error Accumulation

Bin Liu, Wei Sun, Qianqian Wang, Wei Feng, Yijie Chen, Haixi Zhang

CVPR 2026 · Vision-language and multimodal › Cross-modal learning · approach: Unimodal-to-multimodal representation transfer

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 Fourier-based Anchor Alignment (FoA2) (ablation)GSL with GALC · Flickr30K, 40% noiseRSum: 488.4 -> 495.8Improved
add Semantic-Constrained Triplet loss (SCT) (ablation)GSL with GALC and FoA2 · Flickr30K, 40% noiseRSum: 495.8 -> 498.6Improved
apply FoA2 to reduce anchor correlation discrepancy (ablation)GSL · Flickr30K, 20% noiseanchor correlation discrepancyImproved
GSL versus best baseline at 20% noisenoisy correspondence learning · Flickr30K, 20% noiseRSum: 504.7 -> 507.5Improved
GSL versus best baseline at 60% noisenoisy correspondence learning · Flickr30K, 60% noiseRSum: 477.1 -> 484.1Improved
GSL versus best baseline at 20% noise on MS-COCOnoisy correspondence learning · MS-COCO 5-fold, 20% noiseRSum: 524.7 -> 525.1No clear effect
GSL versus best baseline on CC152Knoisy correspondence learning · CC152K (real noise)RSum: 374.2 -> 379.6Improved
set loss weights zeta=1.0 and eta=0.1 for intra- and inter-modal regularizers (ablation)SCT · Flickr30K, 40% noiseRSumImproved

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