List your company

Hallucination Detection on CRAG-MM

0.874AUCEnsemHalDetMIX
0.724240.763120.8020.84088Apr 3, 2026

Evaluation results

MethodLinks
EnsemHalDetMIXVLM Backbone=Qwen2.5-VL-7B2026.040.874
EnsemHalDetAHVLM Backbone=Qwen2.5-VL-7B2026.040.871
MHADattnVLM Backbone=Qwen2.5-VL-7B2026.040.867
EnsemHalDetHSVLM Backbone=Qwen2.5-VL-7B2026.040.864
EnsemHalDetHSVLM Backbone=Llama-3.2-11B2026.040.86
MHADffnVLM Backbone=Qwen2.5-VL-7B2026.040.855
MHADattnVLM Backbone=Llama-3.2-11B2026.040.85
MHADhsVLM Backbone=Qwen2.5-VL-7B2026.040.849
EnsemHalDetMIXVLM Backbone=Llama-3.2-11B2026.040.849
EnsemHalDetAHVLM Backbone=Llama-3.2-11B2026.040.843
MHADhsVLM Backbone=Llama-3.2-11B2026.040.838
MINDVLM Backbone=Qwen2.5-VL-7B2026.040.83
MHADffnVLM Backbone=Llama-3.2-11B2026.040.83
MINDVLM Backbone=Llama-3.2-11B2026.040.822
EnsemHalDetMIXVLM Backbone=Pixtral-12B2026.040.818
SAPLMAVLM Backbone=Llama-3.2-11B2026.040.817
EnsemHalDetHSVLM Backbone=Pixtral-12B2026.040.817
EnsemHalDetAHVLM Backbone=Pixtral-12B2026.040.8
SAPLMAVLM Backbone=Qwen2.5-VL-7B2026.040.797
MHADattnVLM Backbone=Pixtral-12B2026.040.784
MHADffnVLM Backbone=Pixtral-12B2026.040.777
MHADhsVLM Backbone=Pixtral-12B2026.040.763
MINDVLM Backbone=Pixtral-12B2026.040.75
SAPLMAVLM Backbone=Pixtral-12B2026.040.73