| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Composite | HarmonicEval | Kendall-c Tau_c66.2 | 161 | 25d ago | |
| Flickr8k-CF | ITIScore | Kendall-b Correlation (tau_b)56.6 | 145 | 25d ago | |
| Flickr8k Expert | Pomms | Kendall Tau-c (tau_c)60.8 | 114 | 25d ago | |
| Flickr8K Expert (test) | Polos | Kendall tau_c56.4 | 76 | 4mo ago | |
| Nebula | RIGEL | Kendall tau_c57.5 | 66 | 25d ago | |
| Pascal-50S (test) | Polos | HC70 | 66 | 4mo ago | |
| Flickr8K-CF (test) | RefFLEUR | Kendall tau_b38.8 | 65 | 4mo ago | |
| Pascal-50S | DENEB | Accuracy87.8 | 44 | 2mo ago | |
| ICBench Long Caption | Human Score76.57 | 40 | 3mo ago | ||
| FOIL | RIGEL | Accuracy (4-ref)99.2 | 33 | 25d ago | |
| CaptionQA | VISE | GPT Score89.96 | 32 | 29d ago | |
| THumb (test) | CLIP + DN*-ref | tau_c26.2 | 18 | 4mo ago | |
| COMPOSITE (COM) (test) | RefFLEUR | Kendall's tau-c64.2 | 17 | 4mo ago | |
| THumB w/ Human 1.0 | RefCLIP-S | Precision31 | 16 | 4mo ago | |
| THumB w/o Human 1.0 | RefCLIP-S | Precision34 | 16 | 4mo ago | |
| LongCap-Arena (test) | ITIScore | Fluency42.3 | 15 | 3mo ago | |
| MMHE (test) | ITIScore | Fluency Score51.3 | 15 | 3mo ago | |
| MCEval 1.0 (test) | DISCODE | Real Style Score87.8 | 12 | 4mo ago | |
| OID Rated Image Caption | Accuracy62.42 | 7 | 4mo ago | ||
| Abstract-50S | RefPAC-S | Mean Accuracy76.9 | 4 | 4mo ago |