• 0 Posts
  • 2 Comments
Joined 1 year ago
cake
Cake day: June 12th, 2023

help-circle
  • The main reasons why it’s Doom specifically are also because:

    • The game is open-source: https://github.com/id-Software/DOOM/pulls?q=is%3Aopen+is%3Apr This makes it much more doable to port it to other platforms (and to strip out anything not absolutely required to get the first level to run when you run into technological limitations) than when you have to rely on unofficial modding tools.

    • It’s nearly 30 years old and designed for computers with only a few megabytes of memory and for processors of well under 100MHz, which are specs which the majority of modern systems have, even embedded systems. It also renders fully on the CPU and doesn’t require specific hardware like a GPU or a specific graphics chip.

    • Being a first person shooter with 3D-ish visuals it looks a lot more impressive than if you show off a simple game like Pong orTetris or something like that. It has the right balance between performance requirements and impressiveness, and it’s also a game that was very popular in its time and it’s instantly recognisable to a lot of people.


  • It’ll include sources if the sentence structure suggests they should be there, but they’ll also just be built by probabilistic insertion of words.

    I’ve seen attempts of people trying to train a LLM on information with sources. The end result was a model that would still hallucinate false information, and follow it up with a convincing looking source that doesn’t actually exist or a link that just leads to a 404 page. The way current LLMs work makes it impossible for them to mention accurate sources by default as they don’t remember full sentences or even any actual information, but just pick up some underlying patterns.

    Currently the best you can do is letting a LLM come up with search engine queries to find relevant and up to date information for a certain question, and then making it formulate an answer based on what it found and including links to the page(s) it used. The main problem here is that LLMs are not great yet at verifying if a source is accurate, and most people will just take anything that mentions a source as a hard fact without even looking at what the source is.