Startseite Neuigkeiten Neueste Forschungsarbeiten Tutorials Datensätze Wiki SOTA LLM-Modelle GPU-Rangliste Veranstaltungen

Deutsch

Image Retrieval On Flickr30K Cn

Metriken

R@1

R@10

R@5

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Modellname	R@1	R@10	R@5	Paper Title	Repository
R2D2 (ViT-L/14)	84.4	98.4	96.7	CCMB: A Large-scale Chinese Cross-modal Benchmark
InternVL-G-FT	85.9	97.1	98.7	InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks
CN-CLIP (RN50)	66.7	94.1	89.4	Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese
CN-CLIP (ViT-L/14@336px)	84.4	98.7	97.1	Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese
CN-CLIP (ViT-H/14)	83.8	98.6	96.9	Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese
CN-CLIP (ViT-B/16)	79.1	97.4	94.8	Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese
Wukong (ViT-B/32)	67.6	94.2	89.6	Wukong: A 100 Million Large-scale Chinese Cross-modal Pre-training Benchmark
R2D2 (ViT-B)	78.3	97.0	94.6	CCMB: A Large-scale Chinese Cross-modal Benchmark
Wukong (ViT-L/14)	77.4	97.0	94.5	Wukong: A 100 Million Large-scale Chinese Cross-modal Pre-training Benchmark
InternVL-C-FT	85.2	97.0	98.5	InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks
CN-CLIP (ViT-L/14)	82.7	98.6	96.7	Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese

0 of 11 row(s) selected.