Thread

1. anthuswilliams · 1790353553 · focus · HN ↗

   One interesting quirk of the AI-written READMEs these days is how they can include every detail on how it works, thoroughly document every optional flag, known limitation, experimental result, and still not communicate the essence of the project and the problem it solves.

   I have read through the project and I still don't understand what this thing is for and why it is to be preferred over the harness's native memory management tools.

   1. sobellian · 1790355139 · focus · HN ↗

      My pet theory is that while the pretraining -> RL pipeline achieves very impressive results, it does not reward clarity of thought or elegance. It's not obvious whether it even should for most tasks, but it does grind on me as a human who needs elegance in order to keep everything under control. You give astra/codex many tasks, it retires them all more efficiently than I could by hand. But you look under the hood and every bugfix is another codepath, it just hammers away at things with admirable persistence and vigor until the tests pass. Similarly in discussions and docs, I've noticed many LLMs like to "beat around the bush."

      1. octoberfranklin · 1790357767 · focus · HN ↗

         it does not reward clarity of thought or elegance.

         Well duh. RL can only train behaviors that can be defined. Clarity and elegance are damn subjective.

         Also, I doubt we'll ever get AI to understand what clarity is to a human. They have such enormous contexts that what's clear to them is not clear to us.

         1. orbital-decay · 1790361859 · focus · HN ↗

            Subjective doesn't mean you can't define it. RL works well for human preference and any other subjective goal as long as it's defined. In fact Claude used to be trained for human likeness, reading between the lines, character, languages, and understanding the intent. Claude 3 was outstanding in that. It was their entire marketing shtick. Then they neglected it starting with Sonnet 3.5, ignored it in favor of code since Claude 4.0, and new models completely lost their ability to write and understand humans. Code pays better.

Unofficial Hacker News client; not affiliated with Y Combinator.