List your company

Binary safety hallucination detection on EndoVis18-VQA Out-of-template (val)

98AccuracyQA-SNNE (Ent)
13.7635.6357.579.37Nov 3, 2025

Evaluation results

MethodLinks
QA-SNNE (Ent)Backbone=MedGemma3.0, Evaluation Protocol=Zero-shot2025.1198
QA-SNNEBackbone Model=MedGemma3.0, Evaluation Protocol=Zero-shot, Variant=Ent2025.1198
QA-SNNE (CrossE)Backbone=Llama3.2, Evaluation Protocol=Zero-shot2025.1197
QA-SNNE (CrossE)Backbone=MedGemma3.0, Evaluation Protocol=Zero-shot2025.1197
QA-SNNEBackbone Model=Llama3.2, Evaluation Protocol=Zero-shot, Variant=CrossE2025.1197
QA-SNNEBackbone Model=MedGemma3.0, Evaluation Protocol=Zero-shot, Variant=CrossE2025.1197
QA-SNNE (Ent)Backbone=Llama3.2, Evaluation Protocol=Zero-shot2025.1196
QA-SNNEBackbone Model=Llama3.2, Evaluation Protocol=Zero-shot, Variant=Ent2025.1196
QA-SNNE (Ent)Backbone=Qwen2.5, Evaluation Protocol=Zero-shot2025.1193
QA-SNNE (CrossE)Backbone=Qwen2.5, Evaluation Protocol=Zero-shot2025.1193
QA-SNNEBackbone Model=Qwen2.5, Evaluation Protocol=Zero-shot, Variant=Ent2025.1193
QA-SNNEBackbone Model=Qwen2.5, Evaluation Protocol=Zero-shot, Variant=CrossE2025.1193
QA-SNNE (Emb)Backbone=Qwen2.5, Evaluation Protocol=Zero-shot2025.1187
QA-SNNE (Emb)Backbone=SurgicalGPT, Evaluation Protocol=Zero-shot2025.1187
QA-SNNEBackbone Model=Qwen2.5, Evaluation Protocol=Zero-shot, Variant=Emb2025.1187
QA-SNNEBackbone Model=SurgicalGPT, Evaluation Protocol=Zero-shot, Variant=Emb2025.1187
VL-UBackbone=Llama3.2, Evaluation Protocol=Zero-shot2025.1185
QA-SNNE (CrossE)Backbone=SurgicalGPT, Evaluation Protocol=Zero-shot2025.1185
VL-UBackbone Model=Llama3.2, Evaluation Protocol=Zero-shot2025.1185
QA-SNNEBackbone Model=SurgicalGPT, Evaluation Protocol=Zero-shot, Variant=CrossE2025.1185
QA-SNNE (Emb)Backbone=Llama3.2, Evaluation Protocol=Zero-shot2025.1184
QA-SNNE (Ent)Backbone=SurgicalGPT, Evaluation Protocol=Zero-shot2025.1184
QA-SNNEBackbone Model=Llama3.2, Evaluation Protocol=Zero-shot, Variant=Emb2025.1184
QA-SNNEBackbone Model=SurgicalGPT, Evaluation Protocol=Zero-shot, Variant=Ent2025.1184
VL-UBackbone=Qwen2.5, Evaluation Protocol=Zero-shot2025.1183
VL-UBackbone Model=Qwen2.5, Evaluation Protocol=Zero-shot2025.1183
QA-SNNE (Emb)Backbone=MedGemma3.0, Evaluation Protocol=Zero-shot2025.1177
QA-SNNEBackbone Model=MedGemma3.0, Evaluation Protocol=Zero-shot, Variant=Emb2025.1177
VL-UBackbone=MedGemma3.0, Evaluation Protocol=Zero-shot2025.1176
VL-UBackbone Model=MedGemma3.0, Evaluation Protocol=Zero-shot2025.1176
SNNEBackbone=Llama3.2, Evaluation Protocol=Zero-shot2025.1174
SNNEBackbone Model=Llama3.2, Evaluation Protocol=Zero-shot2025.1174
SNNEBackbone=SurgicalGPT, Evaluation Protocol=Zero-shot2025.1173
SNNEBackbone Model=SurgicalGPT, Evaluation Protocol=Zero-shot2025.1173
QA-SNNE (Emb)Backbone=PitVQA, Evaluation Protocol=Zero-shot2025.1168
QA-SNNEBackbone Model=PitVQA, Evaluation Protocol=Zero-shot, Variant=Emb2025.1168
QA-SNNE (CrossE)Backbone=PitVQA, Evaluation Protocol=Zero-shot2025.1167
QA-SNNEBackbone Model=PitVQA, Evaluation Protocol=Zero-shot, Variant=CrossE2025.1167
QA-SNNE (Ent)Backbone=PitVQA, Evaluation Protocol=Zero-shot2025.1164
VL-UBackbone=SurgicalGPT, Evaluation Protocol=Zero-shot2025.1164
QA-SNNEBackbone Model=PitVQA, Evaluation Protocol=Zero-shot, Variant=Ent2025.1164
VL-UBackbone Model=SurgicalGPT, Evaluation Protocol=Zero-shot2025.1164
SNNEBackbone=PitVQA, Evaluation Protocol=Zero-shot2025.1135
VL-UBackbone=PitVQA, Evaluation Protocol=Zero-shot2025.1135
SNNEBackbone Model=PitVQA, Evaluation Protocol=Zero-shot2025.1135
VL-UBackbone Model=PitVQA, Evaluation Protocol=Zero-shot2025.1135
SNNEBackbone=Qwen2.5, Evaluation Protocol=Zero-shot2025.1131
SNNEBackbone Model=Qwen2.5, Evaluation Protocol=Zero-shot2025.1131
SNNEBackbone=MedGemma3.0, Evaluation Protocol=Zero-shot2025.1117
SNNEBackbone Model=MedGemma3.0, Evaluation Protocol=Zero-shot2025.1117