List your company
ResearchFormerly WizWand

Hallucination Detection on HALP

0.9349AUROCGemma3-12B
0.5820280.6736390.765250.856861Mar 5, 2026
Updated 8mo ago

Evaluation results

MethodLinks
Gemma3-12BRepresentation=Query Tokens (QT)2026.030.9349
Molmo-V1-7BRepresentation=Query Tokens (QT)2026.030.9193
Qwen2.5-VL-7BRepresentation=Query Tokens (QT)2026.030.915
Phi4-VL-5.6BRepresentation=Query Tokens (QT)2026.030.9033
LLaVa-Next-8BRepresentation=Query Tokens (QT)2026.030.9026
SmolVLM2-2.2BRepresentation=Query Tokens (QT)2026.030.9014
Llama-3.2-11B-VisionRepresentation=Query Tokens (QT)2026.030.8959
Llama-3.2-11B-VisionRepresentation=Average2026.030.8013
Qwen2.5-VL-7BRepresentation=Average2026.030.7902
Qwen2.5-VL-7BRepresentation=Visual Features (VF)2026.030.7873
Phi4-VL-5.6BRepresentation=Vision Tokens (VT)2026.030.7738
SmolVLM2-2.2BRepresentation=Average2026.030.7715
Llama-3.2-11B-VisionRepresentation=Visual Features (VF)2026.030.7703
Phi4-VL-5.6BRepresentation=Average2026.030.7646
Molmo-V1-7BRepresentation=Average2026.030.763
Llama-3.2-11B-VisionRepresentation=Vision Tokens (VT)2026.030.7377
Gemma3-12BRepresentation=Average2026.030.7347
SmolVLM2-2.2BRepresentation=Visual Features (VF)2026.030.7238
LLaVa-Next-8BRepresentation=Average2026.030.7135
FastVLM-7BRepresentation=Vision Tokens (VT)2026.030.7028
SmolVLM2-2.2BRepresentation=Vision Tokens (VT)2026.030.6894
Molmo-V1-7BRepresentation=Vision Tokens (VT)2026.030.6867
FastVLM-7BRepresentation=Visual Features (VF)2026.030.683
Molmo-V1-7BRepresentation=Visual Features (VF)2026.030.683
Gemma3-12BRepresentation=Visual Features (VF)2026.030.6736
Qwen2.5-VL-7BRepresentation=Vision Tokens (VT)2026.030.6683
FastVLM-7BRepresentation=Average2026.030.6665
LLaVa-Next-8BRepresentation=Vision Tokens (VT)2026.030.627
Phi4-VL-5.6BRepresentation=Visual Features (VF)2026.030.6166
FastVLM-7BRepresentation=Query Tokens (QT)2026.030.6136
LLaVa-Next-8BRepresentation=Visual Features (VF)2026.030.6108
Gemma3-12BRepresentation=Vision Tokens (VT)2026.030.5956