Sometimes AI's contributions may be diffuse, and more akin to the suffusion of a tea-bag, or even a drop of tincture, than the addition of an extra cookie on the plate. In such cases there is usually no distinct section that you can wall off and say 'AI did this'. In spite of this, it may have made a rather modest than overwhelming contribution to the work.
Don't credit the LLM
41–50 of 51 posts
Re: Don't credit the LLM
#42This is a little silly — if I quote a large passage from an encyclopedia (or whatever), I cite the source. If I use an LLM to write some of a paper or report, I’m not writing that part (even if I edit it or rephrase it or whatever) so I’d better credit the LLM in whatever way is most consistent with how its output is being presented in the paper.
I think there is a clear distinction to make in the case of communications. Yes, if you are writing a longer piece of communication and you just verbatim ship whatever the LLM spits out, it is important that you note that in the communication. This obviously applies to deepfakes as well. But this quickly gets slippery. If an LLM generates an initial draft of a paragraph, and I edit it 10 times making mostly stylistic…
> If an LLM generates an initial draft of a paragraph, and I edit it 10 times making mostly stylistic changes, do I need to cite the LLM for that passage?
How much stylistic change removes the citation requirement from large blocks of someone else’s text in an academic paper?
Re: Don't credit the LLM
#43Earlier quoted context omitted.
I don't understand this rebuttal at all. There is a huge difference between having an idea and implementing it yourself, and handing your idea (plus suggestions and data structures or whatever) to other people to implement. In the latter case, the team completing the project gets a lot of credit for the work. That has certainly been the case everywhere I've worked, and I wouldn't have it any other way, even in cases…
So are you saying we should have a credits doc for all major contributors to all software and hardware products, both human and LLM? I'd agree with that, this is not available at the moment for the most popular and profitable products in use currently.
I do credit all contributors to my open-source projects, and I think this is the norm, but to be clear (and I did say this in my original comment), I'm only talking about cases where assigning credit is actually important. There are many cases where it isn't.
Re: Don't credit the LLM
#44Personification an LLM has become a style or trend. Saying "Sol solved this problem for me" sounds great, indicating your honesty in giving credit to LLM. However this is wrong. You are "using" LLM to solve something, just like "using" a calculator to do some computation. Still the computation is done by you, not calculator. When you drove from X to Y, you did it, not the car.
Can we still say things like "GCC eliminates the dead code for us so we can leave it written this way", like we did decades ago, or where are the goalposts?
> When you drove from X to Y, you did it, not the car.
Really? Even if it's a self-driving car where you just punched where you want to go, into an app?
Re: Don't credit the LLM
#45> Why did I need to know that? In a perfect world such a disclaimer would help anyone reading the output by setting expectations. We each have different experiences with this, and different levels of acceptance. Knowing that something is using generated content can help anyone either accept / skim / thoroughly review something. The problem is that we're seeing lots of backlash, often for ideological reasons, or for o…
Why did I need to know that? To tell whether your contribution is compatible with our project's chosen redistribution license, and contribution policies.
Re: Don't credit the LLM
#46There is a difference between crediting the LLM (which i agree is silly) and disclosing the use of an LLM, which I think is important. The author seems to conflate these two. A disclosure is appropriate whenever the reader might feel differently about the result of armed with this information. For example, "this research on the non-addictive nature of tobacco, funded by the tobacco industry". Moreover, not citing the…
Disclosure is appropriate when a reader might feel differently. A reader such as, oh, a judge in a courtroom.
> not citing the LLM is plagiarism
Using an LLM to substantially generate the output rather than just make tweaks to one's own work is plagiarism; not citing the LLM is neglecting to admit to plagiarism.
Re: Don't credit the LLM
#47This is a little silly — if I quote a large passage from an encyclopedia (or whatever), I cite the source. If I use an LLM to write some of a paper or report, I’m not writing that part (even if I edit it or rephrase it or whatever) so I’d better credit the LLM in whatever way is most consistent with how its output is being presented in the paper.
I think there is a clear distinction to make in the case of communications. Yes, if you are writing a longer piece of communication and you just verbatim ship whatever the LLM spits out, it is important that you note that in the communication. This obviously applies to deepfakes as well. But this quickly gets slippery. If an LLM generates an initial draft of a paragraph, and I edit it 10 times making mostly stylistic…
The ideas in the paragraph continue to be plagiarized out of the training data of the LLM, in spite of stylistic changes. The paragraph was essentially composed by an interpolative mashup of the works of unknown, unidentified authors.
Re: Don't credit the LLM
#48I'll definitely credit Isaac for his eloquent, non-slop writing style. Tried to start a related discussion lately (https://news.ycombinator.com/item?id=49027642) more focused on the identity of developers in the age of the agent's second or first authorship.
Re: Don't credit the LLM
#49An LLM is not me. It is capable of doing more "work" but it also doesn't have my mental models and experience. My colleagues know roughly my areas of expertise. If I say "I think that X is happening" they'll assume that is the result of my expertise being applied and react accordingly. If I say "[LLM] thinks X is happening" they'll react differently and I want them to . Just like I might say "I suspect X is happening…
the phrase "[LLM] thinks X is happening" is good chunk of the problem it didn't. it produced an output, and you hopefully invoked some human judgement before believing it was worth sharing, and you are hopefully expecting to be accountable if that judgement was incorrect (whatever the failure mode)
Fine. "The LLM produced a chain of reasoning consistent with a belief that X is happening."
I've found it fairly helpful to anthropomorphize systems to communicate things about their behavior, despite them obviously not being human ("The load balancer thought the West US region was chilling", or "that cluster was unhappy because of bad packet loss"). I don't really have many problems with people taking those statements that I think computers have emotions.
Re: Don't credit the LLM
#50This is a little silly — if I quote a large passage from an encyclopedia (or whatever), I cite the source. If I use an LLM to write some of a paper or report, I’m not writing that part (even if I edit it or rephrase it or whatever) so I’d better credit the LLM in whatever way is most consistent with how its output is being presented in the paper.
I think there is a clear distinction to make in the case of communications. Yes, if you are writing a longer piece of communication and you just verbatim ship whatever the LLM spits out, it is important that you note that in the communication. This obviously applies to deepfakes as well. But this quickly gets slippery. If an LLM generates an initial draft of a paragraph, and I edit it 10 times making mostly stylistic…
If you take an initial draft of a paragraph from someone else’s work and reframe it into your own wording, that is considered plagiarism unless you properly attribute the original source. You should follow the same guidelines for the llm. I agree that you should also make sure you understand and believe what you wrote. None of this is very slippery — understand the ideas you present and don’t pretend to have come up with ideas that originated elsewhere.