this post was submitted on 15 Aug 2024
75 points (97.5% liked)

Technology

58115 readers
4389 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 content.
  3. Be excellent to each another!
  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, to ask if your bot can be added please contact us.
  9. Check for duplicates before posting, duplicates may be removed

Approved Bots


founded 1 year ago
MODERATORS
 

A paper[1] presented in June at the NAACL 2024 conference describes "how to apply large language models to write grounded and organized long-form articles from scratch, with comparable breadth and depth to Wikipedia pages." A "research prototype" version of the resulting "STORM" system is available online and has already attracted thousands of users. This is the most advanced system for automatically creating Wikipedia-like articles that has been published to date.

The authors hail from Monica S. Lam's group at Stanford, which has also published several other papers involving LLMs and Wikimedia projects since 2023 (see our previous coverage: WikiChat, "the first few-shot LLM-based chatbot that almost never hallucinates" – a paper that received the Wikimedia Foundation's "Research Award of the Year" some weeks ago).

Please read the article before commenting. Also, coming right up, another paper creates a structural diagram in comic sans.

top 8 comments
sorted by: hot top controversial new old
[–] [email protected] 27 points 1 month ago* (last edited 1 month ago) (1 children)

Sooooo much fact checking will need to be done if we actually want accurate articles... Soon it will also write the facts and start to blur that line of reality vs trumptalk

[–] [email protected] 6 points 1 month ago (1 children)

As said in the article, a big evaluation criteria of the research was whether it provided a good-enough first draft ("pre-write") for actual editors.

[–] [email protected] 14 points 1 month ago (1 children)

The problem is that a lot of people will use it for the entire process, like the research papers that got published with "as a large language model I don't have access to patient data but I can..." buried inside

[–] [email protected] 1 points 1 month ago (1 children)

Only the bad people who write promotional articles would trust this for the entire thing. Serial article creators know better

[–] [email protected] 1 points 1 month ago (1 children)
[–] [email protected] 1 points 1 month ago

What do you mean, you think long-time article creators don’t understand verifiability policies?

[–] [email protected] -4 points 1 month ago (1 children)

Extremely cool. Perhaps a right step in the direction of hallucination free LLMs?

[–] [email protected] 6 points 1 month ago

It's more like a wrong step towards wikipedia being more full of spam, disinfo, etc.