New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add new clinical benchmark datasets #1163

Open

dcecchini opened this issue Jan 13, 2025 · 3 comments

Labels

⏭️ Next Release

Contributor

dcecchini commented Jan 13, 2025 •

edited

Loading

Datasets to add:

dcecchini added the ⏭️ Next Release label

Collaborator

chakravarthik27 commented Jan 13, 2025

EquityMedQA:

https://springernature.figshare.com/articles/dataset/A_Toolbox_for_Surfacing_Health_Equity_Harms_and_Biases_in_Large_Language_Models/26133973

Collaborator

chakravarthik27 commented Jan 13, 2025

Autonomous Medical Evaluation for Guideline Adherence (AMEGA):

https://github.com/DATEXIS/AMEGA-benchmark/

Collaborator

chakravarthik27 commented Jan 13, 2025 •

edited

Loading

Med-Halt Tests

We can implement this by interchanging the ground truth values.

False Confidence Test (FCT): Providing with false answer for evaluation.
None of the Above Test (Nota): adding NOTA option
Fake Questions Test (FQT): Pick and Change Random Question from Dataset

The following tests are not good, in my opinion.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment