Thread

1. crorella · 1790573511 · focus · HN ↗

   It looks like a lot like how data bases query optimizers work, with the exception that in the paper there is also a learning/memory component that conditions the evaluation of the answer provided by the first model.

   1. mdk2578 · 1790581257 · focus · HN ↗

      [dead]

   2. alansaber · 1790589937 · focus · HN ↗

      Given how LLMs access compressed knowledge from their model weights, the similarities make sense

Unofficial Hacker News client; not affiliated with Y Combinator.