Object Hallucination Detection on FOIL-COCO (test)
99.6AccuracyVC-Inspector-3B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| VC-Inspector-3BEvaluation Mode=Reference-free2025.09 | 99.6 | — | — | |
| FLEUREvaluation Mode=Reference-based2025.09 | 97.3 | — | — | |
| FLEUREvaluation Mode=Reference-free2025.09 | 96.8 | — | — | |
| PAC-SEvaluation Mode=Reference-based2025.09 | 93.5 | — | — | |
| RefCLIP-SNumber of References=42022.05 | 92.6 | — | — | |
| BERT-SNumber of References=42022.05 | 92.1 | — | — | |
| RefCLIP-SNumber of References=12022.05 | 91 | — | — | |
| CIDErNumber of References=42022.05 | 90.6 | — | — | |
| MIDNumber of References=12022.05 | 90.5 | — | — | |
| MIDNumber of References=42022.05 | 90.5 | — | — | |
| PAC-SEvaluation Mode=Reference-free2025.09 | 90.2 | — | — | |
| BERT-SNumber of References=12022.05 | 88.6 | — | — | |
| CLIP-SNumber of References=12022.05 | 87.2 | — | — | |
| CLIP-SNumber of References=42022.05 | 87.2 | — | — | |
| SPICENumber of References=42022.05 | 86.1 | — | — | |
| METEORNumber of References=42022.05 | 85.4 | — | — | |
| BLEU-4Number of References=42022.05 | 82.6 | — | — | |
| CIDErNumber of References=12022.05 | 82.5 | — | — | |
| ROUGE-LNumber of References=42022.05 | 79.3 | — | — | |
| METEORNumber of References=12022.05 | 78.8 | — | — | |
| SPICENumber of References=12022.05 | 75.5 | — | — | |
| ROUGE-LNumber of References=12022.05 | 71.7 | — | — | |
| BLEU-4Number of References=12022.05 | 66.5 | — | — | |
| lengthNumber of References=12022.05 | 50.2 | — | — | |
| lengthNumber of References=42022.05 | 50.2 | — | — | |
| BERT-S2022.05 | — | 88.6 | 92.1 | |
| BLEU-42022.05 | — | 66.5 | 82.6 | |
| CIDEr2022.05 | — | 82.5 | 90.6 | |
| CLIP-S2022.05 | — | 87.2 | 87.2 | |
| length2022.05 | — | 50.2 | 50.2 | |
| METEOR2022.05 | — | 78.8 | 85.4 | |
| MID2022.05 | — | 90.5 | 90.5 | |
| RefCLIP-S2022.05 | — | 91 | 92.6 | |
| ROUGE-L2022.05 | — | 71.7 | 79.3 | |
| SPICE2022.05 | — | 75.5 | 86.1 |