1
$\begingroup$

There was an answer on this question that I thought was really helpful.

In fact, I really need the references provided in that answer, but I don't have enough rep to view deleted answers.

Would someone be able to get those for me?

$\endgroup$

1 Answer 1

10
$\begingroup$

Posts are usually deleted for a reason; we will not provide access to the text of deleted answers.

For this particular case, the "references" are completely useless to you anyway since the answer was plagiarized from the output of an artificial language model (also called "chatbots" or "AIs" recently) and the specific "references" mentioned do not actually exist - the now-deleted answer is the only search result for these exact titles.

$\endgroup$
7
  • 6
    $\begingroup$ This waste of the asker’s time was one of our major reasons for asking our users not to paste chatbot output into answers. $\endgroup$
    – rob Mod
    Feb 17 at 17:20
  • 1
    $\begingroup$ I'm sorry but some of the references do actually exist because I found them... Just didn't write them down so can't pull them up again now I actually have time to read them. Can you please at least extract the references so I can go and read the articles that I can find, then it's up to me to determine if they're useless or not. Just because a LLM pulled them out, doesn't immediately make the answer useless, it just means that it can't be assumed to be true. I understand it's a violation of policy to post that stuff, but you'd be doing me a solid if you could help me out on this occasion $\endgroup$ Feb 17 at 17:41
  • 1
    $\begingroup$ @ScottishTapWater No, none of the references exist. There are papers with similar titles by authors with similar names, but not these exact papers with these exact authors. If you want a bunch of ChatGPT-generated paper titles for some reason you can just ask it yourself. There is nothing valuable here. $\endgroup$
    – ACuriousMind Mod
    Feb 17 at 17:59
  • $\begingroup$ Hmm... I'll take your word for it, but even knowing similar papers by similar authors would have been a useful starting point and I don't think your position as a mod on here qualifies you to tell me what is and isn't useful to my research. Nonetheless, thank you for taking the time to answer my concerns $\endgroup$ Feb 17 at 18:02
  • 1
    $\begingroup$ Out of curiosity, how was it determined that the answer was from a chat-bot? $\endgroup$
    – Kyle Kanos
    Feb 18 at 3:51
  • 1
    $\begingroup$ @KyleKanos Well, the non-existent but plausible-sounding citations are one distinctive marker. Specifically for ChatGPT there are other quirks in the way the answer is written that are recognizable (both by humans who have read a bunch of ChatGPT output and AI classifiers). In general we look also at patterns of user behaviour (e.g. a user posting several detailed answers within an impossibly narrow timespan). $\endgroup$
    – ACuriousMind Mod
    Feb 18 at 11:17
  • $\begingroup$ @ACuriousMind I see, TYVM. $\endgroup$
    – Kyle Kanos
    Feb 19 at 3:41

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .