Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

In 2014, Springer and IEEE had to retract 120 comp-sci papers that were gibberish generated with SCIgen. The problem persists as of May 2021[1], and given the advancements in LLMs, I wouldn't be surprised if things are getting worse...

[1] https://asistdl.onlinelibrary.wiley.com/doi/10.1002/asi.2449...



The Sokal article wasn't retracted and the editors still claim that they didn't make any mistake allowing his nonsense to be published. Can you see the difference?


Shrugs. It's an issue, but not necessarily as extreme as some people make it sound, and not entirely limited to the social sciences (see e.g. the justification of the physicist who approved Igor Bogdanov's thesis back in the day: All these were ideas that could possibly make sense. It showed some originality and some familiarity with the jargon. That's all I ask.).


I agree, in the grand scheme of things it's really nothing. It doesn't really bother me that someone somewhere published some nonsense. It only irks me the persistent defense of these obviously wrong decisions. Just admit you fucked up and move on. If the publishers did just that it would long have been forgotten.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: