AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
2min chapter
Building a Unified NLP Framework at LinkedIn with Huiji Gao - #481
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
The Performance Properties of the Dtax Model
When we use this model for people sace, we obsorved a significant lift on the anly searching success ratei i was also interested in the inference performance in terms of time. How did that compare from before and after? Yes, you good point. The answer is, i am not able to provide this result because the bisly modo is not able to ship into product. So yomay get far enough with the baseline model to ye. For the counter models onmy latency. Is wesing 20 million seconds only? Noar, as i can e imagin it would be more than ten seconds, somthing. Tit a latensy quencer. Ah,
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode