Latent Space: The AI Engineer Podcast cover image

[LIVE] Anthropic Distillation & How Models Cheat (SWE-Bench Dead) | Nathan Lambert & Sebastian Raschka

Latent Space: The AI Engineer Podcast

00:00

Found Issues in Sweebench Verified

swyx details findings: many tasks unsolvable or overly specific, and how leaked training data causes future-looking solutions.

Play episode from 35:34
Transcript

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