Chatbot Software Begins to Face Fundamental Limitations | Quanta Magazine
www.quantamagazine.org
Recent results show that large language models struggle with compositional tasks, suggesting a hard limit to their abilities.
  • mindbleach@sh.itjust.worksEnglish
    1·
    9 days ago

    This is adorably misguided. We tried letting humans write the rules. It didn’t work.

    The recent explosion of neural-network stuff is exciting specifically because it doesn’t rely on us understanding how the fuck it works.