That's been my expectation from the start. We'll need a memory system, an execut... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		Grosvenor 4 months ago \| parent \| context \| favorite \| on: Less is more: Recursive reasoning with tiny networ... That's been my expectation from the start. We'll need a memory system, an executive function/reasoning system as well as some sort of sense integration - auditory, visual, text in the case of LLMs, symbolic probably. A good avenue of research would be to see if you could glue opencyc to this for external "knowledge". LLM's are fundamentally a dead end. Github link: https://github.com/SamsungSAILMontreal/TinyRecursiveModels

taneq 4 months ago | [–]

They’re no more a dead end than your hippocampus is a dead end. They’re just not a complete AGI in and of themselves. They’re a component.

naasking 4 months ago | [–]

LLMs are pretty a central to most next step architectures, like neurosymbolic ones, multimodal ones, etc.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact