Zhenhua Han

icon indicating a website link https://hzhua.github.io

icon company Microsoft Research

Results 1 issues of Zhenhua Han

What are the accurate scores of the task-level Radar graph

![Image](https://longbench2.github.io/static/images/radar.png) Could please share the accurate scores (before normalization) for the radar graph in the leaderboard? This will help people to compare the task-level performance with these models. Thanks