SOTAVerified

Cross-Modal Retrieval

Cross-Modal Retrieval (CMR) is a task of retrieving items across different modalities, such as image, text, video, and audio. The core challenge of CMR is the heterogeneity gap, which arises because data from different modalities have distinct representations, making direct comparison difficult. To address this, most CMR methods focus on learning a shared latent embedding space. In this space, concepts from different modalities are projected, allowing their similarity to be measured using a distance metric.

Scene-centric vs. Object-centric Image-Text Cross-modal Retrieval: A Reproducibility Study

Papers

Showing 110 of 522 papers

TitleStatusHype
An analysis of vision-language models for fabric retrieval0
Mask-aware Text-to-Image Retrieval: Referring Expression Segmentation Meets Cross-modal Retrieval0
Maximal Matching Matters: Preventing Representation Collapse for Robust Cross-Modal Retrieval0
Multimodal Medical Image Binding via Shared Text Embeddings0
FedNano: Toward Lightweight Federated Tuning for Pretrained Multimodal Large Language Models0
ContextRefine-CLIP for EPIC-KITCHENS-100 Multi-Instance Retrieval Challenge 2025Code0
SA-Person: Text-Based Person Retrieval with Scene-aware Re-ranking0
FOLIAGE: Towards Physical Intelligence World Models Via Unbounded Surface Evolution0
EmotionRankCLAP: Bridging Natural Language Speaking Styles and Ordinal Speech Emotion via Rank-N-Contrast0
DocMMIR: A Framework for Document Multi-modal Information RetrievalCode0
Show:102550
← PrevPage 1 of 53Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1MaMMUT (ours)Image-to-text R@170.7Unverified
2VASTText-to-image R@168Unverified
3X2-VLM (large)Text-to-image R@167.7Unverified
4BEiT-3Text-to-image R@167.2Unverified
5XFM (base)Text-to-image R@167Unverified
6X2-VLM (base)Text-to-image R@166.2Unverified
7PTP-BLIP (14M)Text-to-image R@164.9Unverified
8OmniVL (14M)Text-to-image R@164.8Unverified
9VSE-GradientText-to-image R@163.6Unverified
10X-VLM (base)Text-to-image R@163.4Unverified
#ModelMetricClaimedVerifiedStatus
1X2-VLM (large)Image-to-text R@198.8Unverified
2X2-VLM (base)Image-to-text R@198.5Unverified
3BEiT-3Image-to-text R@198Unverified
4OmniVL (14M)Image-to-text R@197.3Unverified
5Aurora (ours, r=128)Image-to-text R@197.2Unverified
6ERNIE-ViL 2.0Image-to-text R@197.2Unverified
7X-VLM (base)Image-to-text R@197.1Unverified
8VSE-GradientImage-to-text R@197Unverified
9ALIGNImage-to-text R@195.3Unverified
10VASTText-to-image R@191Unverified
#ModelMetricClaimedVerifiedStatus
1VLPCook (R1M+)Image-to-text R@174.9Unverified
2VLPCookImage-to-text R@173.6Unverified
3T-Food (CLIP)Image-to-text R@172.3Unverified
4T-FoodImage-to-text R@168.2Unverified
5X-MRSImage-to-text R@164Unverified
6H-TImage-to-text R@160Unverified
7SCANImage-to-text R@154Unverified
8ACMEImage-to-text R@151.8Unverified
9VLPCookImage-to-text R@145.2Unverified
10AdaMineImage-to-text R@139.8Unverified
#ModelMetricClaimedVerifiedStatus
1HarMA (w/ GeoRSCLIP)Mean Recall38.95Unverified
2GeoRSCLIP-FTMean Recall38.87Unverified
3GLISAMean Recall37.69Unverified
4RemoteCLIPMean Recall36.35Unverified
5PE-RSITR (MRS-Adapter)Mean Recall31.12Unverified
6PIRMean Recall24.46Unverified
7DOVEMean Recall22.72Unverified
8SWANMean Recall20.61Unverified
9GaLRMean Recall18.96Unverified
10AMFMNMean Recall15.53Unverified
#ModelMetricClaimedVerifiedStatus
1HarMA (w/ GeoRSCLIP)Image-to-text R@132.74Unverified
2GeoRSCLIP-FTImage-to-text R@132.3Unverified
3GLISAImage-to-text R@132.08Unverified
4RemoteCLIPImage-to-text R@128.76Unverified
5PE-RSITR (MRS-Adapter)Image-to-text R@123.67Unverified
6PIRImage-to-text R@118.14Unverified
7DOVEImage-to-text R@116.81Unverified
8GaLRImage-to-text R@114.82Unverified
9SWANImage-to-text R@113.35Unverified
10AMFMNImage-to-text R@110.63Unverified
#ModelMetricClaimedVerifiedStatus
1CLASS (ORMA)Hits@167.4Unverified
2ORMAHits@166.5Unverified
3Song et al.Hits@156.5Unverified
4CLASS (AMAN)Hits@151.1Unverified
5DSOKRHits@151Unverified
6AMANHits@149.4Unverified
7All-EnsembleHits@134.4Unverified
8MLP1Hits@122.4Unverified
9GCN2Hits@122.3Unverified
#ModelMetricClaimedVerifiedStatus
1NAPRegImage-to-text R@181.9Unverified
2Dual-path CNNImage-to-text R@141.2Unverified
#ModelMetricClaimedVerifiedStatus
1ResNet-18Median Rank565Unverified
2GeoCLAPMedian Rank159Unverified
#ModelMetricClaimedVerifiedStatus
1Dual PathText-to-image Medr2Unverified
#ModelMetricClaimedVerifiedStatus
1NAPRegImage-to-text R@156.2Unverified
#ModelMetricClaimedVerifiedStatus
13SHNetImage-to-text R@185.8Unverified
#ModelMetricClaimedVerifiedStatus
1NAPRegText-to-image R@143Unverified