5min chapter

The Data Scientist Show cover image

Making LLMs hallucinate less, how to diagnose ML models, from PM in Google AI to CEO of Galileo - Vikram Chatterji - The Data Scientist Show #066

The Data Scientist Show

CHAPTER

How to Diagnose and Detect Hallucination Using Metrics

The metrics you mentioned using the token level probability it called it a gray box measurement because it's not completely black box. The second thing becomes trying to identify context and especially for you know a business use cases you have a lot of your own context. A big bottleneck amongst when you talk to customers is these vector databases are really created being databases but they're not necessarily providing you with the right kind of context. You can go and fetch just the rightkind of context and and supplement that so that the model now knows exactly what you're talking about.

00:00

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

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

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