LLM are entirely the wrong tool for that. They’ve used neural nets for their anti cheat for a while but cheating has nothing to do with language so large language models are not for that. They could however use them to police the steam forums but only if a comment gets reported otherwise you’re putting every comment through an llm which would cost a fuckton
You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: [email protected]
No game suggestions, friend requests, surveys, or begging.
No Let’s Plays, streams, highlight reels/montages, random videos or shorts.
No off-topic posts/comments, within reason.
Use the original source, no clickbait titles, no duplicates.
(Submissions should be from the original source if possible, unless from paywalled or non-english sources.
If the title is clickbait or lacks context you may lightly edit the title.)
Seems like they could be experimenting with an LLM to improve server-side anti-cheat.
LLM are entirely the wrong tool for that. They’ve used neural nets for their anti cheat for a while but cheating has nothing to do with language so large language models are not for that. They could however use them to police the steam forums but only if a comment gets reported otherwise you’re putting every comment through an llm which would cost a fuckton