some problem

[dpcdsg]

It’s not like the author didn’t provide details on how to calculate scores. The evaluation only generates dictionary results？

[CPunisher]

The score is simply the number of passes divided by the total number of samples and averaged.

Raw data: https://github.com/java-bench/JavaBench/blob/main/paper_plot/visual_table_context.ipynb
Division: https://github.com/java-bench/JavaBench/blob/main/paper_plot/visual_table.py#L45