AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
#132 - FraudGPT, Apple GPT, unlimited jailbreaks, RT-2, Frontier Model Forum, PhotoGuard
Last Week in AI
How to Use Images and Sounds for Indirect Instruction Injection in Multi Model Lems
3min Snip
00:00
Summary
Transcript
Episode notes
Do you see model distillation as a way to get around the size issue, or do you see something more fundamental there? The language text inputs are sort of key to being able to reason about instructions and have all this knowledge encoded in it. So when you distill things, you inherently lose some of that reasoning capability and that knowledge. To some extent, probably, but also probably not to the extent that it's needed for edge hardware.
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