The Gradient: Perspectives on AI cover image

Kyunghyun Cho: Neural Machine Translation, Language, and Doing Good Science

The Gradient: Perspectives on AI

00:00

Is There a Benchmark for Evaluating Models?

The evaluation component of that is really interesting. It seems that many of the questions that the community is trying to think about right now, not just a distributional hypothesis, don't seem to have just yet evaluation benchmarks. So what that means is that the metrics must be also often learned. And then the dynamic nature has to be supported or facilitated by human annotators as well.

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