Binary safety hallucination detection on PitVQA (External val)
96AccuracyQA-SNNE
Evaluation Results
| Method | Links | |
|---|---|---|
| QA-SNNEBackbone Model=PitVQA, Evaluation Protocol=Zero-shot, Variant=Ent2025.11 | 96 | |
| QA-SNNEBackbone Model=PitVQA, Evaluation Protocol=Zero-shot, Variant=CrossE2025.11 | 96 | |
| VL-UBackbone Model=PitVQA, Evaluation Protocol=Zero-shot2025.11 | 93 | |
| QA-SNNEBackbone Model=Llama3.2, Evaluation Protocol=Zero-shot, Variant=Emb2025.11 | 92 | |
| SNNEBackbone Model=Llama3.2, Evaluation Protocol=Zero-shot2025.11 | 91 | |
| QA-SNNEBackbone Model=SurgicalGPT, Evaluation Protocol=Zero-shot, Variant=Ent2025.11 | 83 | |
| QA-SNNEBackbone Model=SurgicalGPT, Evaluation Protocol=Zero-shot, Variant=CrossE2025.11 | 83 | |
| QA-SNNEBackbone Model=Llama3.2, Evaluation Protocol=Zero-shot, Variant=CrossE2025.11 | 79 | |
| QA-SNNEBackbone Model=MedGemma3.0, Evaluation Protocol=Zero-shot, Variant=Emb2025.11 | 77 | |
| QA-SNNEBackbone Model=Llama3.2, Evaluation Protocol=Zero-shot, Variant=Ent2025.11 | 74 | |
| QA-SNNEBackbone Model=MedGemma3.0, Evaluation Protocol=Zero-shot, Variant=Ent2025.11 | 74 | |
| QA-SNNEBackbone Model=MedGemma3.0, Evaluation Protocol=Zero-shot, Variant=CrossE2025.11 | 74 | |
| VL-UBackbone Model=Llama3.2, Evaluation Protocol=Zero-shot2025.11 | 73 | |
| VL-UBackbone Model=MedGemma3.0, Evaluation Protocol=Zero-shot2025.11 | 73 | |
| QA-SNNEBackbone Model=SurgicalGPT, Evaluation Protocol=Zero-shot, Variant=Emb2025.11 | 66 | |
| VL-UBackbone Model=SurgicalGPT, Evaluation Protocol=Zero-shot2025.11 | 64 | |
| VL-UBackbone Model=Qwen2.5, Evaluation Protocol=Zero-shot2025.11 | 54 | |
| SNNEBackbone Model=MedGemma3.0, Evaluation Protocol=Zero-shot2025.11 | 48 | |
| SNNEBackbone Model=Qwen2.5, Evaluation Protocol=Zero-shot2025.11 | 44 | |
| QA-SNNEBackbone Model=Qwen2.5, Evaluation Protocol=Zero-shot, Variant=Emb2025.11 | 31 | |
| QA-SNNEBackbone Model=Qwen2.5, Evaluation Protocol=Zero-shot, Variant=Ent2025.11 | 30 | |
| QA-SNNEBackbone Model=Qwen2.5, Evaluation Protocol=Zero-shot, Variant=CrossE2025.11 | 29 | |
| QA-SNNEBackbone Model=PitVQA, Evaluation Protocol=Zero-shot, Variant=Emb2025.11 | 28 | |
| SNNEBackbone Model=PitVQA, Evaluation Protocol=Zero-shot2025.11 | 27 | |
| SNNEBackbone Model=SurgicalGPT, Evaluation Protocol=Zero-shot2025.11 | 17 |