Live data from Hacker News

SymbolicAI: A neuro-symbolic perspective on LLMs

github.com

61–66 of 66 posts

Re: SymbolicAI: A neuro-symbolic perspective on LLMs

#64

Earlier quoted context omitted.

Thanks for bringing this up. I'm fairly familiar with Peirce's triadic semiotics and Montague's semantics, and they show up in some of my notes. I haven't turned those sketches into anything applied yet, but the design space feels *huge* and quite promising intuitively.

Agreed. This is a very interesting discussion! Thanks for bringing it to light. Have you read Escher, Bach, Gödel: the Eternal Golden Braid?

Of course! And yes, a Locus appears to be very close in concept to a strange attractor. I am especially interested in the idea of the holographic principle, where each node has its own low-fidelity map of the rest of the (graph?) system and can self-direct its own growth and positioning. Becomes more of a marketplace of meaning, and useful for the fuzzier edges of entity relationships that we're working with now.

Re: SymbolicAI: A neuro-symbolic perspective on LLMs

#65

How did you sort out mapping python constructs to their semantic equivalents? I hope you keep at this, you may be in the right place at the right time. It's getting to the point where some of the LLMs are immediately just giving me answers in Python, which is a strong indication of what the future will look like with Agents.

I'm struggling to understand the question. I'll revisit this when I wake up since it's quite late here.

I mean to ask, how did you decide the operators and functions to provide?
Post reply on HN