203
ChatGPT gets code questions wrong 52% of the time (www.theregister.com)
submitted 3 years ago by to c/technology@lemmy.ml
66 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] 21 points 3 years ago* (4 children)

Wait a second here... I skimmed the paper and GitHub and didn't find an answer to a very important question: is this GPT3.5 or 4? There's a huge difference in code quality between the two and either they made a giant accidental omission or they are being intentionally misleading. Please correct me if I missed where they specified that. I'm assuming they were using GPT3.5, so yeah those results would be as expected. On the HumanEval benchmark, GPT4 gets 67% and that goes up to 90% with reflexion prompting. GPT3.5 gets 48.1%, which is exactly what this paper is saying. (source).

  • permalink
  • fedilink
  • source
  • hideshow 8 child comments
    •
  • [–] 5 points 3 years ago (1 child)

    ChatGPT is 3.5, 4 is just called GPT4

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 4 points 3 years ago (1 child)

    Hmm that's incorrect. ChatGPT (if you pay for it) does both.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 7 points 3 years ago (1 child)

    I'm talking about the models and how they're written about in the literature. I don't care how OpenAI brands their products.

    From the paper itself:

    For the additional 2000 SO questions, ChatGPT 3.5 Turbo API is used.

    https://arxiv.org/pdf/2308.02312.pdf

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 2 points 3 years ago (1 child)

    Oh ok! Got it. I read it as you saying ChatGPT doesn't use GPT 4. It's still unclear what they used for part of it because of the bit before the part you quoted:

    For each of the 517 SO questions, the first two authors manually used the SO question’s title, body, and tags to form one question prompt3 and fed that to the Chat Interface [45] of ChatGPT.

    It doesn't say if it's 4 or 3.5, but I'm going to assume 3.5. Anyway, in the end they got the same result for GPT 3.5 that it gets on HumanEval, which isn't anything interesting. Also, GPT 4 is much better, so I'm not really sure what the point is. Their stuff on the analysis of the language used in the questions was pretty interesting though.

    Also, thanks for finding their mention of 3.5. I missed that in my skim through obviously.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 2 points 3 years ago (1 child)

    For sure, no worries. I had the same questions as you when reading it. Fwiw, the paper is really kind of sloppy. I think it's maybe a case of poor students not wanting to pay for GPT-4? Maybe they'll clean it up and respond to some of the criticisms when it comes out of draft, but it doesn't seem like very rigorous scholarship to me.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [–] 2 points 3 years ago

    Yeah I think you're right on about the students not being able to afford GPT4 (I don't blame them. The API version gets expensive quick). I agree though that it doesn't seem super well put together.

  • permalink
  • fedilink
  • source
  • parent
  • [–] 3 points 3 years ago (1 child)

    Whatever GitHub Copilot uses (the version with the chat feature), I don't find its code answers to be particularly accurate. Do we know which version that product uses?

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •