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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add Fourier-based Anchor Alignment (FoA2) (ablation) | GSL with GALC · Flickr30K, 40% noise | RSum: 488.4 -> 495.8 | Improved |
| add Semantic-Constrained Triplet loss (SCT) (ablation) | GSL with GALC and FoA2 · Flickr30K, 40% noise | RSum: 495.8 -> 498.6 | Improved |
| apply FoA2 to reduce anchor correlation discrepancy (ablation) | GSL · Flickr30K, 20% noise | anchor correlation discrepancy | Improved |
| GSL versus best baseline at 20% noise | noisy correspondence learning · Flickr30K, 20% noise | RSum: 504.7 -> 507.5 | Improved |
| GSL versus best baseline at 60% noise | noisy correspondence learning · Flickr30K, 60% noise | RSum: 477.1 -> 484.1 | Improved |
| GSL versus best baseline at 20% noise on MS-COCO | noisy correspondence learning · MS-COCO 5-fold, 20% noise | RSum: 524.7 -> 525.1 | No clear effect |
| GSL versus best baseline on CC152K | noisy correspondence learning · CC152K (real noise) | RSum: 374.2 -> 379.6 | Improved |
| set loss weights zeta=1.0 and eta=0.1 for intra- and inter-modal regularizers (ablation) | SCT · Flickr30K, 40% noise | RSum | Improved |
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.
- BiCro — BiCro: Noisy Correspondence Rectification for Multi-Modality Data via Bi-Directional Cross-Modal Similarity Consistency (CVPR 2023) · better on 1 of 1 shared result
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
- Mitigating Noisy Correspondence by Geometrical Structure Consistency Learning (CVPR 2024 · similarity 0.91)
- Robust Noisy Correspondence Learning with Equivariant Similarity Consistency (CVPR 2024 · similarity 0.89)
- Noisy Correspondence Rectification via Asymmetric Similarity Learning (AAAI 2025 · similarity 0.88)
- Noisy Correspondence Learning with Self-Reinforcing Errors Mitigation (AAAI 2024 · similarity 0.87)
- Learning to Rematch Mismatched Pairs for Robust Cross-Modal Retrieval (CVPR 2024 · similarity 0.87)
- Cross-Modal and Uni-Modal Soft-Label Alignment for Image-Text Retrieval (AAAI 2024 · similarity 0.86)
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.
