Image Captioning Hallucination Detection on FOIL (test)
95.4AccuracyPolos
Evaluation Results
| Method | Links | |
|---|---|---|
| Polosnum_references=4-ref2024.02 | 95.4 | |
| RefPAC-Snum_references=4-ref2024.02 | 94.9 | |
| RefPAC-Snum_references=1-ref2024.02 | 93.7 | |
| Polosnum_references=1-ref2024.02 | 93.3 | |
| RefCLIP-Snum_references=4-ref2024.02 | 92.6 | |
| BERTScorenum_references=4-ref2024.02 | 92.1 | |
| BARTScorenum_references=4-ref2024.02 | 91.1 | |
| RefCLIP-Snum_references=1-ref2024.02 | 91 | |
| CIDErnum_references=4-ref2024.02 | 90.6 | |
| MIDnum_references=1-ref2024.02 | 90.5 | |
| MIDnum_references=4-ref2024.02 | 90.5 | |
| PAC-Snum_references=1-ref2024.02 | 89.9 | |
| PAC-Snum_references=4-ref2024.02 | 89.9 | |
| BERTScorenum_references=1-ref2024.02 | 88.6 | |
| MoverScorenum_references=1-ref2024.02 | 88.4 | |
| MoverScorenum_references=4-ref2024.02 | 88.4 | |
| CLIP-Snum_references=1-ref2024.02 | 87.2 | |
| CLIP-Snum_references=4-ref2024.02 | 87.2 | |
| SPICEnum_references=4-ref2024.02 | 86.1 | |
| BARTScorenum_references=1-ref2024.02 | 85.3 | |
| BLEUnum_references=4-ref2024.02 | 82.6 | |
| METEORnum_references=4-ref2024.02 | 82.6 | |
| CIDErnum_references=1-ref2024.02 | 82.5 | |
| ROUGEnum_references=4-ref2024.02 | 79.3 | |
| METEORnum_references=1-ref2024.02 | 78.8 | |
| SPICEnum_references=1-ref2024.02 | 75.5 | |
| ROUGEnum_references=1-ref2024.02 | 71.7 | |
| BLEUnum_references=1-ref2024.02 | 66.5 |