←back to thread

Over fifty new hallucinations in ICLR 2026 submissions

(gptzero.me)
504 points puttycat | 1 comments | 07 Dec 25 13:16 UTC | HN request time: 0s | source
Show context
theoldgreybeard ◴[07 Dec 25 15:09 UTC] No.46182214[source]▶
If a carpenter builds a crappy shelf “because” his power tools are not calibrated correctly - that’s a crappy carpenter, not a crappy tool.

If a scientist uses an LLM to write a paper with fabricated citations - that’s a crappy scientist.

AI is not the problem, laziness and negligence is. There needs to be serious social consequences to this kind of thing, otherwise we are tacitly endorsing it.

replies(37): >>46182289 #>>46182330 #>>46182334 #>>46182385 #>>46182388 #>>46182401 #>>46182463 #>>46182527 #>>46182613 #>>46182714 #>>46182766 #>>46182839 #>>46182944 #>>46183118 #>>46183119 #>>46183265 #>>46183341 #>>46183343 #>>46183387 #>>46183435 #>>46183436 #>>46183490 #>>46183571 #>>46183613 #>>46183846 #>>46183911 #>>46183917 #>>46183923 #>>46183940 #>>46184450 #>>46184551 #>>46184653 #>>46184796 #>>46185025 #>>46185817 #>>46185849 #>>46189343 #
thaumasiotes ◴[07 Dec 25 15:31 UTC] No.46182401[source]▶
> If a scientist uses an LLM to write a paper with fabricated citations - that’s a crappy scientist.

Really? Regardless of whether it's a good paper?

replies(2): >>46182457 #>>46182500 #
zwnow ◴[07 Dec 25 15:38 UTC] No.46182457[source]▶
How is it a good paper if the info in it cant be trusted lmao
replies(1): >>46182610 #
thaumasiotes ◴[07 Dec 25 15:55 UTC] No.46182610[source]▶
Whether the information in the paper can be trusted is an entirely separate concern.

Old Chinese mathematics texts are difficult to date because they often purport to be older than they are. But the contents are unaffected by this. There is a history-of-math problem, but there's no math problem.

replies(3): >>46182787 #>>46183475 #>>46187709 #
1. alexcdot ◴[08 Dec 25 02:38 UTC] No.46187709[source]▶
Problem is that most ML papers today are not independently verifiable proofs - in most, you have to trust the scientist didn't fraudulently produce their results.

There is so much BS being submitted to conferences and decreasing the amount of BS they see would result in less skimpy reviews and also less apathy