List your company

Hallucination Evaluation on AMBER Generative Task

67.1CoverageGPT-4V
36.83644.69352.5560.407Apr 28, 2026May 2, 2026May 6, 2026May 11, 2026May 15, 2026May 19, 2026May 24, 2026

Evaluation results

MethodLinks
GPT-4VBase Model=Proprietary2026.0567.130.72.64.6
MFPOBase Model=LLaVA-v1.5-13B2026.0556.119.41.43.4
MFPOBase Model=LLaVA-v1.5-7B2026.0555.722.51.94.1
AMP-MEGBase Model=LLaVA-v1.5-13B2026.0553.845.85.611
LBPBase Model=LLaVA-v1.5-7B2026.0553.218.51.63.5
HALVABase Model=LLaVA-v1.5-13B2026.0552.630.43.26.4
mDPOBase Model=LLaVA-v1.5-7B2026.0552.424.52.44.4
LBPBase Model=LLaVA-v1.5-13B2026.0551.516.61.33.3
LLaVA-v1.5-7BBase Model=LLaVA-v1.5-7B2026.0550.539.14.68.5
RLAIF-VBase Model=LLaVA-v1.5-7B2026.0550.316.113
LLaVA-v1.5-13BBase Model=LLaVA-v1.5-13B2026.0550.237.34.38.8
V-DPOBase Model=LLaVA-v1.5-7B2026.0549.727.32.75.6
OctopusVLM=LLaVA-1.52026.0449.223.41.2—
OctopusVLM=InstructBLIP2026.0448.522.21.3—
OPA-DPOBase Model=LLaVA-v1.5-7B2026.0547.911.60.92.2
HSA-DPOBase Model=LLaVA-v1.5-13B2026.0547.313.41.22.1
RLHF-VBase Model=LLaVA-v1.5-13B2026.0546.125.12.16.3
CCA-LLaVABase Model=LLaVA-v1.5-7B2026.0545.932.14.18.1
VCDVLM=InstructBLIP2026.044531.82.6—
ICDVLM=LLaVA-1.52026.0444.5312.2—
RegularVLM=InstructBLIP2026.0444.431.82.6—
ICDVLM=InstructBLIP2026.0442.4312.2—
VCDVLM=LLaVA-1.52026.0441.536.33.1—
IECD2VLM=LLaVA-1.52026.0440.55.60.4—
IECD2VLM=InstructBLIP2026.0439.610.40.5—
RegularVLM=LLaVA-1.52026.043844.531—