Correlation with Human Judgment on Flickr8K-CF
46Tau BVC-Inspector-7B
Evaluation Results
| Method | Links | |
|---|---|---|
| VC-Inspector-7BReference Setting=Reference-free2025.09 | 46 | |
| VC-Inspector-3BReference Setting=Reference-free2025.09 | 39 | |
| FLEURReference Setting=Reference-based2025.09 | 38.8 | |
| PAC-S++Reference Setting=Reference-based, Backbone=ViT-L2025.09 | 38.8 | |
| FLEURReference Setting=Reference-free2025.09 | 38.6 | |
| PAC-S++Reference Setting=Reference-free, Backbone=ViT-L2025.09 | 38.5 | |
| HICE-SReference Setting=Reference-based2025.09 | 38.2 | |
| PAC-S++Reference Setting=Reference-based, Backbone=ViT-B2025.09 | 37.9 | |
| Poloscategory=Learning-based metrics2024.02 | 37.8 | |
| PolosReference Setting=Reference-based2025.09 | 37.8 | |
| RefPAC-Scategory=Learning-based metrics2024.02 | 37.6 | |
| PAC-SReference Setting=Reference-based2025.09 | 37.6 | |
| MIDcategory=Similarity-based metrics2024.02 | 37.3 | |
| HICE-SReference Setting=Reference-free2025.09 | 37.2 | |
| PAC-S++Reference Setting=Reference-free, Backbone=ViT-B2025.09 | 37 | |
| RefCLIP-Sreferences=with ground-truth2021.04 | 36.4 | |
| RefCLIP-Scategory=Similarity-based metrics2024.02 | 36.4 | |
| CLIPScoreReference Setting=Reference-based2025.09 | 36.4 | |
| PAC-Scategory=Learning-based metrics2024.02 | 36 | |
| PAC-SReference Setting=Reference-free2025.09 | 36 | |
| CLIP-Sreferences=none2021.04 | 34.4 | |
| CLIP-Scategory=Similarity-based metrics2024.02 | 34.4 | |
| CLIPScoreReference Setting=Reference-free2025.09 | 34.4 | |
| UMICcategory=Learning-based metrics2024.02 | 30.1 | |
| LEIC2021.04 | 29.5 | |
| LEICcategory=Similarity-based metrics2024.02 | 29.5 | |
| CIDEr2021.04 | 24.6 | |
| CIDErcategory=Classic metrics2024.02 | 24.6 | |
| CIDErReference Setting=Reference-based2025.09 | 24.6 | |
| SPICE2021.04 | 24.4 | |
| SPICEcategory=Classic metrics2024.02 | 24.4 | |
| SPICEReference Setting=Reference-based2025.09 | 24.4 | |
| BARTScorecategory=Similarity-based metrics2024.02 | 24.3 | |
| BERT-Sbackbone=RoBERTa-F2021.04 | 22.8 | |
| MoverScorecategory=Similarity-based metrics2024.02 | 22.8 | |
| BERTScorecategory=Similarity-based metrics2024.02 | 22.8 | |
| BERTScoreReference Setting=Reference-based2025.09 | 22.8 | |
| METEOR2021.04 | 22.2 | |
| METEORcategory=Classic metrics2024.02 | 22.2 | |
| METEORReference Setting=Reference-based2025.09 | 22.2 | |
| ROUGE-L2021.04 | 19.9 | |
| ROUGEcategory=Classic metrics2024.02 | 19.9 | |
| ROUGEReference Setting=Reference-based2025.09 | 19.9 | |
| BLEU_1Reference Setting=Reference-based2025.09 | 17.9 | |
| BLEU-42021.04 | 16.9 | |
| BLEU_4Reference Setting=Reference-based2025.09 | 16.9 | |
| BLEUcategory=Classic metrics2024.02 | 16.4 | |
| SPARCScategory=Classic metrics2024.02 | 10.4 |