NousResearch/atropos
1
0
Fork 0
mirror of https://github.com/NousResearch/atropos.git synced 2026-04-30 17:40:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Add wandb

Browse source
This commit is contained in:
Dylan Anderson 2025-05-18 18:00:21 -07:00
parent 444bd5b1d7
commit 7e91a94a3e
1 changed files with 3 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

3
environments/hack0/doctor_agent/readme.md
View file

@ -1,6 +1,9 @@
Persona-Aware MedQA Benchmarking
https://youtube.com/shorts/02GEURik0PQ
Wandb: https://wandb.ai/nous-hackathon-2/atropos-environments_hack0_doctor_agent?nw=nwusertsadpbb
We intended on adding a simple percentage accurate score but couldn't get it done in time :(
In this project, we reimagined medical QA evaluation by introducing a persona filter—a novel layer that simulates real-world variability in patient communication styles. Leveraging the MedQA dataset as our foundation, we infused each scenario with distinct personas generated via xAI’s language models:
1. The Cooperative Patient – open, verbose, and highly informative.