News for hackers. Stuff that matters.

A comment on GPTZero finds 100 new hallucinations in NeurIPS 2025 accepted papers. Parent comment

Re:GPTZero finds 100 new hallucinations in NeurIPS 2025 acce... by nativeit on Thursday January 22, 2026 @02:21PM (#46723913)
I see your point, but I don’t see where the author makes any claims about the specifics of the hallucinations, or their impact on the papers’ broader validity. Indeed, I would have found the removal of supposed “innocuous” examples to be far more deceptive than simply calling a spade a spade, and allowing the data to speak for itself.
[ Parent | on HN ]
Re:GPTZero finds 100 new hallucinations in NeurIPS 2025 acce... by gowld on Thursday January 22, 2026 @04:02PM (#46725092)
The point is that they should focus on the meaningful errors, not the automiation of meaningless errors.
[ Parent | on HN ]
Re:GPTZero finds 100 new hallucinations in NeurIPS 2025 acce... by reliabilityguy on Thursday January 22, 2026 @04:38PM (#46725465)
Why these are meaningless? How do I know now that the whole paper is not a slop?
[ Parent | on HN ]
Re:GPTZero finds 100 new hallucinations in NeurIPS 2025 acce... by j2kun on Thursday January 22, 2026 @04:31PM (#46725383)
The author calls the mistakes "confirmed hallucinations" without proof (just more or less evidence). The data never "speak for itself." The author curates the data and crafts a story about it. This story presented here is very suggestive (even using the term "hallucination" is suggestive). But calling it "100 suspected hallucinations", or "25 very likely hallucinations" does less for the author's end goal: selling their service.
[ Parent | on HN ]
Re:GPTZero finds 100 new hallucinations in NeurIPS 2025 acce... by gold23 on Thursday January 22, 2026 @06:36PM (#46726539)
Obviously a post on a startup's blog will be more editorialized than an academic paper. Still, this seems like an important discussion to have.
[ Parent | on HN ]