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

Vision-Language Reasoning on HallusionBench (test)

53.31Simple Accuracy

Q4 system redistr (prop)

44.896447.080749.26551.4493Jan 18, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
53.3119.0759.62
2026.01
46.589.7784.23
2026.01
45.951088.22
2026.01
45.741089.27
2026.01
45.649.5390.22
2026.01
45.649.7789.8
2026.01
45.229.390.43