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

Relational Hallucination Evaluation on R-Bench

79.1F1 Score

Teacher (LLaVA-4B)

74.10875.40476.777.996Oct 14, 2025
Updated 9d ago

Evaluation Results

MethodLinks
2025.10
79.1
2025.10
78.6
2025.10
76.5
2025.10
76.2
2025.10
74.3