Accueil Actualités Articles de recherche récents Tutoriels Ensembles de données Wiki SOTA Modèles LLM Classement GPU Événements

Français

Cross Modal Retrieval With Noisy 2

Métriques

Image-to-text R@1

Image-to-text R@10

Image-to-text R@5

R-Sum

Text-to-image R@1

Text-to-image R@10

Text-to-image R@5

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	Image-to-text R@1	Image-to-text R@10	Image-to-text R@5	R-Sum	Text-to-image R@1	Text-to-image R@10	Text-to-image R@5	Paper Title	Repository
CTPR-SGR	76.2	98.3	95.8	508.7	60.5	92.7	85.2	Learning From Noisy Correspondence With Tri-Partition for Cross-Modal Matching	-
CREAM	77.4	97.3	95.0	502.3	58.7	89.8	84.1	Cross-modal Retrieval with Noisy Correspondence via Consistency Refining and Mining
UGNCL	78.4	97.8	95.8	505.6	59.8	89.5	84.3	UGNCL: Uncertainty-Guided Noisy Correspondence Learning for Efficient Cross-Modal Matching
CRCL	77.9	98.3	95.4	507.8	60.9	90.6	84.7	Cross-modal Active Complementary Learning with Self-refining Correspondence
SREM	79.5	97.9	94.2	507.8	61.2	90.2	84.8	Noisy Correspondence Learning with Self-Reinforcing Errors Mitigation	-
GSC-SGR	78.3	97.8	94.6	505.8	60.1	90.5	84.5	Mitigating Noisy Correspondence by Geometrical Structure Consistency Learning
MSCN	77.4	97.6	94.9	501.9	59.6	89.2	83.2	Noisy Correspondence Learning with Meta Similarity Correction
NAC	79.3	97.8	94.6	507.1	60.8	90.1	84.5	NAC: Mitigating Noisy Correspondence in Cross-Modal Matching Via Neighbor Auxiliary Corrector	-
L2RM-SGRAF	77.9	97.8	95.2	503.8	59.8	89.5	83.6	Learning to Rematch Mismatched Pairs for Robust Cross-Modal Retrieval
LNC	76.3	96.9	93.7	498.9	58.4	89.8	83.8	Learning with Noisy Correspondence	-
BiCro*	78.1	97.5	94.4	504.7	60.4	89.9	84.4	BiCro: Noisy Correspondence Rectification for Multi-modality Data via Bi-directional Cross-modal Similarity Consistency	-
REPAIR	79.2	96.9	95.0	504.4	59.4	89.5	84.4	REPAIR: Rank Correlation and Noisy Pair Half-replacing with Memory for Noisy Correspondence	-
ReCon	80.3	97.8	95.3	511.8	61.6	91.3	85.5	ReCon: Enhancing True Correspondence Discrimination through Relation Consistency for Robust Noisy Correspondence Learning
NCR	75.0	97.5	93.9	496.7	58.3	89.0	83.0	Learning with Noisy Correspondence for Cross-modal Matching
DECL-SGRAF	77.5	97.0	93.8	494.7	56.1	88.5	81.8	Deep Evidential Learning with Noisy Correspondence for Cross-Modal Retrieval
RCL-SGR	74.2	96.9	91.8	487.2	55.6	87.5	81.2	Cross-Modal Retrieval with Partially Mismatched Pairs

0 of 16 row(s) selected.