Thread

1. jmull · 1790477369 · focus · HN ↗

   There's no point to try to adapt our languages to the strengths of LLMs when the strength of LLMs is working in terms of our languages.

   Implement whatever abstractions you think LLMs should work in terms of in whatever language is handy, and have your LLM use those abstractions.

   1. hollars · 1790485718 · focus · HN ↗

      I think I get your point but I think it's reductionist to the point of being incorrect. LLMs must be better at some semantics than others. Programming languages don't have random semantics, they have what matches the world and what matches our languages and so on. And the current frontier LLMs aren't so generic that they can predict any phrase no matter the quality of the content and grammar. Concretely I mean some languages are harder to reason about (predict) than others.

      1. jmull · 1790513301 · focus · HN ↗

         What semantics do you guess LLMs would better at, the semantics they are trained on, or something new with altered semantics?

         1. hollars · 1790518202 · focus · HN ↗

            The semantics they are trained on include latent representations of the world that are superior to any pre-existing PL corpus. That is my guess and it's not any less rigorous than your guess.

Unofficial Hacker News client; not affiliated with Y Combinator.