5 ms·

Yup. Is there a good/forgiving "drunken JSON parser" library that people like to use? Feels like it would be a useful (and separable) piece?

• golol 3y ago
  Honestly, I suspect asking GPT-4 to fix your JSON (in a new chat) is a good drunken JSON parser. We are only scraping the surface of what's possible with LLMs. If Token generation was free and instant we could come up with a giant schema of interacting model calls that generates 10 suggestions, iterates over them, ranks them and picks the best one, as silly as it sounds.
  • hhh 3y ago
    I already do this today to create domain-specific knowledge focused prompts and then have them iterate back and forth and a ‘moderator’ that chooses what goes in and what doesn’t.
  • andai 3y ago
    That's hilarious... if parsing GPT's JSON fails, keep asking GPT to fix it until it parses!
    • golol 3y ago
      It shouldn't be surprising though. If a human makes an error parsing JSON, what do you do? You make them look over it again. Unless their intelligence is the bottleneck they might just be able to fix it.
    • golergka 3y ago
      It works. Just be sure to build a good error message.