List your company

Hallucination Detection on MMMU Pro

88.9AUCMHADhs
60.40467.80275.282.598Apr 3, 2026

Evaluation results

MethodLinks
MHADhsVLM Backbone=Llama-3.2-11B2026.0488.9
MHADffnVLM Backbone=Llama-3.2-11B2026.0487.4
MHADattnVLM Backbone=Llama-3.2-11B2026.0487
MINDVLM Backbone=Llama-3.2-11B2026.0486.6
EnsemHalDetHSVLM Backbone=Llama-3.2-11B2026.0486
EnsemHalDetAHVLM Backbone=Llama-3.2-11B2026.0485.7
EnsemHalDetMIXVLM Backbone=Llama-3.2-11B2026.0485.4
SAPLMAVLM Backbone=Llama-3.2-11B2026.0485
EnsemHalDetHSVLM Backbone=Pixtral-12B2026.0482.6
MHADattnVLM Backbone=Pixtral-12B2026.0482.2
MHADffnVLM Backbone=Pixtral-12B2026.0481
EnsemHalDetAHVLM Backbone=Pixtral-12B2026.0480.2
EnsemHalDetMIXVLM Backbone=Pixtral-12B2026.0480.2
MINDVLM Backbone=Pixtral-12B2026.0480.1
MHADhsVLM Backbone=Pixtral-12B2026.0479.2
SAPLMAVLM Backbone=Pixtral-12B2026.0479.1
EnsemHalDetHSVLM Backbone=Qwen2.5-VL-7B2026.0476.2
MHADattnVLM Backbone=Qwen2.5-VL-7B2026.0475.1
EnsemHalDetMIXVLM Backbone=Qwen2.5-VL-7B2026.0474.8
MHADffnVLM Backbone=Qwen2.5-VL-7B2026.0474.2
EnsemHalDetAHVLM Backbone=Qwen2.5-VL-7B2026.0474
MHADhsVLM Backbone=Qwen2.5-VL-7B2026.0470.7
MINDVLM Backbone=Qwen2.5-VL-7B2026.0468.6
SAPLMAVLM Backbone=Qwen2.5-VL-7B2026.0461.5