Models Are Getting Dumber on Purpose

(w4g1.dev)

42 points | by hruvhwe 54 minutes ago

5 comments

  • kennywinker 5 minutes ago
    Ideally what I'd like to see is pluggable knowledge bases.

    So if I'm e.g. coding a SwiftUI app for navigation, I'd take 9B of basic coding and reasoning, add 10B of swift/swiftUI, add 5B of GIS/geography knowledge and another 5B of frontend app design knowledge. My model doesn't need to know a single line of python.

    Then when I want to research electronics components, I grab a 15B model of agentic research techniques, and add in 10B of electronics knowledge, etc.

    I don't want general purpose models. They try to be everything to everyone. I want to click together a model that is laser-focused on what I am doing, and I want to run it locally

    • jvuygbbkuurx 1 minute ago
      If that is better (and possible) it will be baked in the tools.
    • Gecko4072 3 minutes ago
      Sounds like unix philosophy. Or like Neo downloading Kung-Fu. Good either way.
  • msdz 1 minute ago
    Great article.

    > There's a version of this future where the model card stops listing a knowledge cutoff at all, because what's left in the weights goes stale on a scale of years instead of weeks.

    Future?

    Even just recently I’ve read of two approaches to this problem:

    Needle [0][1] have come up with their tool-calling focused 14 MB model (still an LLM!) – no world knowledge engrained.

    And instead of say, tool call structure, VibeThinker [2][3] focuses on reasoning over world knowledge.

    Combine these two approaches with a reliable search tool/a safe way of accessing the internet for the model, and you’ve got a probably slightly slower model for factual questions, which on the upside however doesn’t hallucinate.

    [0] https://cactuscompute.com/needle

    [1] https://news.ycombinator.com/item?id=49246804

    [2] https://arxiv.org/abs/2606.16140

    [3] https://news.ycombinator.com/item?id=48639240

  • kaufmann 8 minutes ago
    I think the idea is reasonable, however the SimpleQA Bench stopped measuring in Sept. 2025.

    So newer data would be interesting.

    (It seems a bit like an AI generated argument that uses old facts - something that happens to me quite often)

  • Gecko4072 7 minutes ago
    With all the focus on coding and agentic use, I wonder if the rest of the world will notice or care? Most AI use is not for coding or harnesses contrary to what this website thinks. Maybe this is how American companies stay winning.

    For example I prefer Kimi K2.6 1T parameter to Flash V4 0731 230B parameter, even if it is less intelligent.

    • kennywinker 3 minutes ago
      What is the rest of the world using LLMs for? Agentic stuff seems pretty universal. Obviously coding stuff is only interesting to people who want to code, but automating complex digital tasks seems useful in all kinds of contexts.
      • Espressosaurus 0 minutes ago
        I’m working at a company where everyone is using LLMs for everything and I’m not aware of anybody using anything truly genetic. It’s all human in the loop babysitting the tool.
      • Gecko4072 0 minutes ago
        Learning new things for example, general curiosities, text processing
  • stefan_ 5 minutes ago
    I hope there is interesting research being done to stop models from 9B to 3T from authoritatively giving the wrong answers to factual questions instead of doing tool calls, but this article will tell you nothing about it. That's because it's clearly prompted by someone who simply noticed this deficiency very obvious to all practitioners and even users.