Spaces:
Running
on
CPU Upgrade
Is there a downloadable version of the leaderboard results as CSV format?
I want to further compare the results but it is hard to copy and paste the results from screen.
Hi,
You can find the downloadable summaries at - https://github.com/vectara/hallucination-leaderboard?tab=readme-ov-file#data
Hi,
You can find the downloadable summaries at - https://github.com/vectara/hallucination-leaderboard?tab=readme-ov-file#data
Thanks for your reply,
@viveksourabh
But this is not the leaderboard shown on the main page. Do you have the one that already aggregates the evaluation results (just the same as the main page leaderboard but in CSV format)?
Hey
@zhiminy
,
Sorry for the late response. However, we don't have a downloadable version of the aggregated eval results.
Is there an historic data of this? so for example I can compare how GPT 4 has improved within time?
Hi @Agoncalves85 , there has been no re-evaluation of the models. therefore, we cannot track how models improve over time. but this is a good suggestion and we will consider in the future.