List your company

Hallucination Evaluation on POPE Popular MSCOCO, A-OKVQA, GQA average

88.48AccuracyVanilla
80.99282.93684.8886.824Feb 14, 2026

Evaluation results

MethodLinks
VanillaMLLM=Qwen3-VL2026.0288.4886.6191.3588.8
PAIMLLM=Qwen3-VL2026.0288.4585.7892.688.92
AdaVBoostMLLM=Qwen3-VL2026.0288.3586.6191.0488.64
VGAMLLM=Qwen3-VL2026.0287.4684.9791.2987.9
VAFMLLM=Qwen3-VL2026.0287.1483.4593.1187.88
VanillaMLLM=LLaVA-NeXT2026.0286.0181.8393.8287.15
AdaVBoostMLLM=LLaVA-NeXT2026.0286.0181.8593.887.14
VAFMLLM=LLaVA-NeXT2026.0284.6478.8695.8986.33
PAIMLLM=LLaVA-NeXT2026.0284.0978.8395.1385.88
VGAMLLM=LLaVA-NeXT2026.0283.4578.9493.2285.12
VGAMLLM=InternVL3.52026.0283.3177.3194.585
VanillaMLLM=InternVL3.52026.0282.8577.692.5984.35
VAFMLLM=InternVL3.52026.0282.8476.4595.1784.74
AdaVBoostMLLM=InternVL3.52026.0282.6476.3294.9184.55
PAIMLLM=InternVL3.52026.0281.2875.1893.5383.31