List your company

Hallucination Mitigation on HallusionBench visual-dependent setting (test)

33.21qAcc (Question Accuracy)ReVisiT
9.560415.700221.8427.9798Jun 11, 2025

Evaluation results

MethodLinks
ReVisiTBackbone=InternVL3-8B2025.0633.2139.1359.3966.0752.19
ReVisiTBackbone=Qwen2.5-VL-7B2025.0632.8536.0961.7664.2656.93
DoLaBackbone=InternVL3-8B2025.0632.4937.3959.965.751.82
SIDBackbone=Qwen2.5-VL-7B2025.0632.1336.0961.5965.755.11
SIDBackbone=InternVL3-8B2025.0631.0536.0958.0461.7352.19
M3IDBackbone=Qwen2.5-VL-7B2025.0630.3233.4860.0762.8254.38
GreedyBackbone=Qwen2.5-VL-7B2025.0629.9634.7861.0865.753.65
VCDBackbone=Qwen2.5-VL-7B2025.0629.635.2261.2564.6254.74
CODEBackbone=Qwen2.5-VL-7B2025.0628.8831.7457.5357.7654.74
VCDBackbone=InternVL3-8B2025.0628.8835.6557.3662.8249.64
CODEBackbone=InternVL3-8B2025.0628.5236.5256.6862.0949.27
M3IDBackbone=InternVL3-8B2025.0628.1633.9157.0260.6550.73
DoLaBackbone=Qwen2.5-VL-7B2025.0627.0830.8758.0462.0951.46
GreedyBackbone=InternVL3-8B2025.0625.9933.9156.3561.7348.54
ReVisiTBackbone=LLaVA-1.5-7B2025.0620.2226.5251.4445.1354.38
VCDBackbone=LLaVA-1.5-7B2025.0614.4422.6147.5544.7747.45
DoLaBackbone=LLaVA-1.5-7B2025.0613.3622.6149.2445.8549.27
SIDBackbone=LLaVA-1.5-7B2025.0612.2721.348.2244.7748.18
M3IDBackbone=LLaVA-1.5-7B2025.0611.9123.0448.2243.6849.27
GreedyBackbone=LLaVA-1.5-7B2025.0611.5522.6148.2244.448.54
CODEBackbone=LLaVA-1.5-7B2025.0610.4720.4346.3641.5247.81