How shall we perform an evaluation on this benchmark? How do we submit to the leaderboard?
#2
by
GJZhang
- opened
Thanks for the awesome work!
We wish to conduct an evaluation on this benchmark. However, we didn't find any code to perform evaluation codes. May I know how to leverage this benchmark and do an evaluation of an MLLM model? And how shall we submit to your leaderboard?
Hi there, we will release the code shortly (after the Neurips review period). Alternatively, if you model has a publicly accessible API, please kindly email it to [email protected] and we could evaluate it for you. Also, there is a submission section on our website: https://dolphin-sound.github.io/u2-bench/