HyperAI超神経

ホームニュース最新論文チュートリアルデータセット百科事典 SOTA LLMモデル GPU ランキング学会

サイトについて

日本語

HyperAI超神経

Cross Modal Retrieval With Noisy 1

評価指標

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

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名	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
REPAIR	40.5	76.1	67.7	369.2	40.3	76.4	68.2	REPAIR: Rank Correlation and Noisy Pair Half-replacing with Memory for Noisy Correspondence	-
BiCro*	40.8	76.1	67.2	370.2	42.1	76.4	67.6	BiCro: Noisy Correspondence Rectification for Multi-modality Data via Bi-directional Cross-modal Similarity Consistency	-
DECL-SGRAF	39.0	75.5	66.1	364.3	40.7	76.7	66.3	Deep Evidential Learning with Noisy Correspondence for Cross-Modal Retrieval
UGNCL	43.6	74.9	67.1	373.1	42.7	76.4	68.4	UGNCL: Uncertainty-Guided Noisy Correspondence Learning for Efficient Cross-Modal Matching
NAC	41.8	77.3	68.6	373.5	40.5	77.0	68.3	NAC: Mitigating Noisy Correspondence in Cross-Modal Matching Via Neighbor Auxiliary Corrector	-
MSCN	40.1	76.6	65.7	366.7	40.6	76.3	67.4	Noisy Correspondence Learning with Meta Similarity Correction
RCL-SGRAF	41.7	73.6	66.0	364.4	41.6	75.1	66.4	Cross-Modal Retrieval with Partially Mismatched Pairs
ReCon	43.1	78.1	68.7	380.5	44.9	77.4	68.3	ReCon: Enhancing True Correspondence Discrimination through Relation Consistency for Robust Noisy Correspondence Learning
L2RM-SGRAF	43.0	75.7	67.5	374.2	42.8	77.2	68.0	Learning to Rematch Mismatched Pairs for Robust Cross-Modal Retrieval
CRCL	41.8	76.5	67.4	373.7	41.6	78.4	68.0	Cross-modal Active Complementary Learning with Self-refining Correspondence
NCR	39.5	73.5	64.5	355.6	40.3	73.2	64.6	Learning with Noisy Correspondence for Cross-modal Matching
SREM	40.9	77.1	67.5	372.2	41.5	77.0	68.2	Noisy Correspondence Learning with Self-Reinforcing Errors Mitigation	-
GSC-SGR	42.1	77.7	68.4	375.1	42.2	77.1	67.6	Mitigating Noisy Correspondence by Geometrical Structure Consistency Learning
CREAM	40.3	77.1	68.5	372.6	40.2	78.3	68.2	Cross-modal Retrieval with Noisy Correspondence via Consistency Refining and Mining
LNC	39.5	73.1	64.0	355.5	40.6	73.5	64.8	Learning with Noisy Correspondence	-

0 of 15 row(s) selected.