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

Human Evaluation on ScienceVQA (test)

8.35Faithfulness Score

MMBoundary

4.62685.59346.567.5266May 29, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.05
8.357.468.027.94
2025.05
7.556.927.127.19
2025.05
7.497.186.286.98
2025.05
6.815.976.346.37
2025.05
6.745.616.46.23
4.775.244.534.85