Data Skeptic cover image

Debiasing GPT-3 Job Ads

Data Skeptic

00:00

How to Measure Realism in the Final Analysis?

The method was I would say good for our purposes that could be improved upon. So essentially we had humans, so humans meaning people in our research group and friends that were familiar with the work annotate some of the real job ads we gathered. And then realism in the final analysis is the probability of the model saying, hey, this is a real job ad, right?

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app