Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Multimodal Hallucination Assessment on HallusionBench

70Accuracy

Qwen3.5-27B

46.360852.497958.63564.7721Apr 9, 2026Apr 14, 2026Apr 19, 2026Apr 24, 2026Apr 29, 2026May 4, 2026May 9, 2026
Updated 15d ago

Evaluation Results

MethodLinks
2026.04
70
2026.04
67.4
2026.04
66.7
2026.04
63.7
2026.04
63.2
2026.05
54.88
2026.05
54.77
2026.05
54.22
2026.05
53.35
2026.05
53.35
2026.05
52.42
2026.05
52.35
2026.05
52.25
2026.05
52.25
2026.05
51.83
2026.05
51.76
2026.05
50.55
2026.05
48.46
2026.05
48.42
2026.05
48.31
2026.05
47.27