Image Captioning on MSCOCO (val) (CS, Ci, Recall)
53CS ScoreBaseline
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| BaselineModel=LLaVA-1.5-7B2026.04 | 53 | 18.4 | 67 | |
| VAFModel=LLaVA-1.5-7B2026.04 | 53 | 17.3 | 69.9 | |
| VCDModel=LLaVA-1.5-7B2026.04 | 51 | 17.5 | 70.1 | |
| OPERAModel=LLaVA-1.5-7B2026.04 | 49 | 16 | 70.2 | |
| Active-LookModel=LLaVA-1.5-7B2026.04 | 15 | 15.8 | 64.9 |