List your company

Hallucination Evaluation on HaloQuest

80.8Score SQwen2.5-VL + FINER-Tuning
34.93646.84358.7570.657Mar 15, 2026Mar 16, 2026Mar 17, 2026Mar 18, 2026

Evaluation results

MethodLinks
Qwen2.5-VL + FINER-TuningSize=7B2026.0380.8——
Qwen2.5-VLSize=7B2026.0374.8——
InternVL-3.5 + FINER-TuningSize=8B2026.0373.5——
InternVL-3.5 + FINER-TuningSize=14B2026.0371——
InternVL-3.5Size=14B2026.0365——
LLaVA-1.6 + FINER-TuningSize=7B2026.0363.5——
OmniLMM + RLAIF-VSize=12B2026.0362.4——
InternVL-3.5Size=8B2026.0362.4——
Qwen2.5-VL-7B+VRGABackbone=Qwen2.5-VL-7B, Enhancement=VRGA2026.0354.973.9657.8
Qwen2.5-VL-7BBackbone=Qwen2.5-VL-7B2026.0350.266.6759.5
Qwen2.5-VL-3B+VRGABackbone=Qwen2.5-VL-3B, Enhancement=VRGA2026.0348.859.0340.5
Qwen3-VL-30B+VRGABackbone=Qwen3-VL-30B, Enhancement=VRGA2026.0346.568.0574.4
Qwen3-VL-30BBackbone=Qwen3-VL-30B2026.0345.469.276.2
Qwen2.5-VL-3BBackbone=Qwen2.5-VL-3B2026.0344.558.8760.1
LLaVA-1.6Size=7B2026.0344.2——
Qwen2-VL-7BBackbone=Qwen2-VL-7B2026.0340.653.8162.2
Qwen2-VL-7B+VRGABackbone=Qwen2-VL-7B, Enhancement=VRGA2026.0340.454.4760.8
OmniLMMSize=12B2026.0339.9——
Qwen2-VL-7B+ICOTBackbone=Qwen2-VL-7B, Enhancement=ICOT2026.0336.747.8563.8
Qwen2-VL-7B+CCOTBackbone=Qwen2-VL-7B, Enhancement=CCOT2026.03—29.3—
Qwen2.5-VL-3B+CCOTBackbone=Qwen2.5-VL-3B, Enhancement=CCOT2026.03—41.06—
Qwen2.5-VL-7B+CCOTBackbone=Qwen2.5-VL-7B, Enhancement=CCOT2026.03—51.08—