this post was submitted on 11 Mar 2026
55 points (98.2% liked)

Technology

82518 readers
3942 users here now

This is a most excellent place for technology news and articles.

Our Rules

  1. Follow the lemmy.world rules.
  2. Only tech related news or articles.
  3. Be excellent to each other!
  4. Mod approved content bots can post up to 10 articles per day.
  5. Threads asking for personal tech support may be deleted.
  6. Politics threads may be removed.
  7. No memes allowed as posts, OK to post as comments.
  8. Only approved bots from the list below, this includes using AI responses and summaries. To ask if your bot can be added please contact a mod.
  9. Check for duplicates before posting, duplicates may be removed
  10. Accounts 7 days and younger will have their posts automatically removed.

Approved Bots

founded 2 years ago
MODERATORS
L3s@lemmy.world
enu@lemmy.world
technopagan@lemmy.world
L4s@lemmy.world
L3s@hackingne.ws
55
How Much Do LLMs Hallucinate in Document Q&A Scenarios? A 172-Billion-Token Study Across Temperatures, Context Lengths, and Hardware Platforms [TLDR: 25%] (arxiv.org)
submitted 13 hours ago* (last edited 13 hours ago) by RandAlThor@lemmy.ca to c/technology@lemmy.world
25 comments fedilink hide all child comments
 

Evaluating 35 open-weight models across three context lengths (32K, 128K, 200K), four temperatures, and three hardware platforms—consuming 172 billion tokens across more than 4,000 runs—we find that the answer is “substantially, and unavoidably.” Even under optimal conditions—best model, best temperature, temperature chosen specifically to minimize fabrication—the floor is non-zero and rises steeply with context length. At 32K, the best model (GLM 4.5) fabricates 1.19% of answers, top-tier models fabricate 5–7%, and the median model fabricates roughly 25%.

you are viewing a single comment's thread
view the rest of the comments
[–] unpossum@sh.itjust.works -2 points 11 hours ago (1 children)

GLM 4.5 is from August. Isn’t the real tl;dr that a seven month old open model, which was behind proprietary models at the time, did better than most humans would?

[–] MHard@lemmy.world 3 points 8 hours ago (1 children)

The task described in this article is asking questions about a document that was provided to the llm in the context.

I would hope that if you give a human a text and ask them to cite facts from it they would do better than 99% correct.

Also, when the tokens exceeded 200k, the llm error rate was higher than 10%

[–] unpossum@sh.itjust.works 1 points 45 minutes ago

I would hope that if you give a human a text and ask them to cite facts from it they would do better than 99% correct.

That’s literally what school exams are about, isn’t it?

Token window is a problem for all llms though, that’s not easily solved, but it can be worked around to a certain extent.