Papers › Learning with Noisy Correspondence

Learning with Noisy Correspondence

13 Apr 2024International Journal of Computer Vision 2024 4archive 2025-07-28

Zhenyu Huang, Peng Hu, guocheng niu, Xinyan Xiao, Jiancheng Lv, Xi Peng

This paper studies a new learning paradigm for noisy labels, i.e., noisy correspondence (NC). Unlike the well-studied noisy labels that consider the errors in the category annotation of a sample, the NC refers to the errors in the alignment relationship of two data points. Although such false positive pairs are common especially in the data harvested from the Internet, which however are neglected by most existing works. By taking cross-modal retrieval as a showcase, we propose a method called learning with noisy correspondence (LNC). In brief, the LNC first roughly obtains the clean and noisy subsets from the original data and then rectifies the false positive pairs by using a novel adaptive prediction function. Finally, the LNC adopts a novel triplet loss with soft margins to endow cross-modal retrieval the robustness to the NC. To verify the effectiveness of the proposed LNC, we conduct experiments on six benchmark datasets in image-text and video-text retrieval tasks. Besides the effectiveness of the LNC, the experimental results show the necessity of the explicit solution to the NC faced by not only the standard model training paradigm but also the pre-training and fine-tuning paradigms.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Cross-Modal RetrievalCross-modal retrieval with noisy correspondenceRetrievalText RetrievalVideo-Text Retrieval

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Cross-modal retrieval with noisy correspondence CC152K LNC Image-to-text R@1 39.5 #15 of 15 Archive leaderboard report
Cross-modal retrieval with noisy correspondence CC152K LNC Image-to-text R@10 73.1 #15 of 15 Archive leaderboard report
Cross-modal retrieval with noisy correspondence CC152K LNC Image-to-text R@5 64.0 #15 of 15 Archive leaderboard report
Cross-modal retrieval with noisy correspondence CC152K LNC R-Sum 355.5 #15 of 15 Archive leaderboard report
Cross-modal retrieval with noisy correspondence CC152K LNC Text-to-image R@1 40.6 #15 of 15 Archive leaderboard report
Cross-modal retrieval with noisy correspondence CC152K LNC Text-to-image R@10 73.5 #15 of 15 Archive leaderboard report
Cross-modal retrieval with noisy correspondence CC152K LNC Text-to-image R@5 64.8 #15 of 15 Archive leaderboard report
Cross-modal retrieval with noisy correspondence COCO-Noisy LNC Image-to-text R@1 78.2 #14 of 17 Archive leaderboard report
Cross-modal retrieval with noisy correspondence COCO-Noisy LNC Image-to-text R@10 98.5 #14 of 17 Archive leaderboard report
Cross-modal retrieval with noisy correspondence COCO-Noisy LNC Image-to-text R@5 95.8 #14 of 17 Archive leaderboard report
Cross-modal retrieval with noisy correspondence COCO-Noisy LNC R-Sum 519.9 #14 of 17 Archive leaderboard report
Cross-modal retrieval with noisy correspondence COCO-Noisy LNC Text-to-image R@1 62.6 #14 of 17 Archive leaderboard report
Cross-modal retrieval with noisy correspondence COCO-Noisy LNC Text-to-image R@10 95.4 #14 of 17 Archive leaderboard report
Cross-modal retrieval with noisy correspondence COCO-Noisy LNC Text-to-image R@5 89.4 #14 of 17 Archive leaderboard report
Cross-modal retrieval with noisy correspondence Flickr30K-Noisy LNC Image-to-text R@1 76.3 #13 of 16 Archive leaderboard report
Cross-modal retrieval with noisy correspondence Flickr30K-Noisy LNC Image-to-text R@10 96.9 #13 of 16 Archive leaderboard report
Cross-modal retrieval with noisy correspondence Flickr30K-Noisy LNC Image-to-text R@5 93.7 #13 of 16 Archive leaderboard report
Cross-modal retrieval with noisy correspondence Flickr30K-Noisy LNC R-Sum 498.9 #13 of 16 Archive leaderboard report
Cross-modal retrieval with noisy correspondence Flickr30K-Noisy LNC Text-to-image R@1 58.4 #13 of 16 Archive leaderboard report
Cross-modal retrieval with noisy correspondence Flickr30K-Noisy LNC Text-to-image R@10 89.8 #13 of 16 Archive leaderboard report
Cross-modal retrieval with noisy correspondence Flickr30K-Noisy LNC Text-to-image R@5 83.8 #13 of 16 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Triplet Loss

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections