Hacker News Reader

Comment by ambrozk | Hacker News Reader

The paper does things like ask GPT-4 to write code and then check if that code compiles. Since March, they've fine-tuned GPT-4 to add back-ticks around code, which improves human-readable formatting but stops the code compiling. This is interpreted as "degraded performance" in the paper even though it's improved performance from a human perspective.

View on HN

There is degraded performance because GPT4 refuses to carry out certain tasks. To figure it out though, you must need to be able to switch between GPT4-0316 and GPT4-0614. The task it is reluctant to do include:

- legal advice

- psychological guidance

- complex programming tasks.

IMO OpenAI is just backtracking on what it released to resegment their product into multiple offerings.

>IMO OpenAI is just backtracking on what it released to resegment their product into multiple offerings.

I think this is it, but also it is pulling back on the value of products if it reduces their compute costs.

I'm hoping competition will be so fierce in this market that quality and opaque changes to priced LLM experiences won't be a thing for very long.

Hilarious because Google is taking the exact opposite approach.

They released Bard as a terrible model to begin with, and now you can program and have it interpret images. It's been consistently improving.

OpenAI really got sidetracked when they put out such a good model that freaked everyone out. Google saw this and decided to do the opposite to prevent too much attention. Now Google can improve quietly and nobody will notice.

I think it's probably a good idea that GPT4 avoids legal or psychological tasks. Those are areas where giving incorrect output can have catastrophic consequences, and I can see why GPT4's developers want to avoid potential liability.

Yes, considering those are fields where humans have to be professionally educated and licensed, and also carry liabilty for any mistakes. It probably shouldn't be used for civil or mechanical engineering either.

That's right, no serious work should be done with it. But programming is fine.

https://xkcd.com/2030/

I wasn't asking for advice. I was in fact exploring ways to measure employee performance (using GPT as a search engine) and criticized one of the assumptions in one of theoretical framework with my own experience. GPT retorted "burnout", I retorted "harassment grounded on material evidence", and it spitted a generic block about how it couldn't provide legal or psychological guidance. I wasn't asking for advice, it was just a point in a wider conversation. Switching to GPT-4-0314 it proceeded with a reply that matched the discussion main's topic (metrics in HR) switching it back to GPT-4-0613 it outputted the exact same generic block.

It's ridiculous. We should have access to a completely neutered version if we want it, we should just have to sign for access if they are worried about begin sued.

The query explicitly asks it to add no other text to the code.

> it's improved performance from a human perspective.

Ignoring explicit requirements is the kind of thing that makes modern day search engines a pain to use.

If I'm using it from the web UI, this is exactly what I would want—this allows the language model to define the output language so I get correct syntax highlighting, without an error-prone secondary step of language detection.

If I'm using it from the API, then all I have to do is strip out the leading backticks and language name if I don't need to check the language, or alternatively parse it to determine what the output language is.

It seems to me that in either case this is actually strictly better, and annotating the computer programming language used doesn't feel to me like extra text—I would think of that requirement as prohibiting a plaintext explanation before or after the code.

It is still ignoring an explicit requirement, which is almost always bad. The user should be able to override what the creator/application thinks is 'strictly better'. Exceptions probably exist, but this isn't one of them.

Like I said, I read their requirement differently—the phrase they use is "the code only".

I don't think that including backticks is a violation of this requirement. It's still readily parseable and serves as metadata for interpreting the code. In the context of ChatGPT, which typically will provide a full explanation for the code snippet, I think this is a reasonable interpretation of the instruction.

I understand your point better now. I'm still not sure how I feel about it, because code + metadata is still not "the code only", but it's not a totally unreasonable interpretation of the phrase.

This sounds a lot like a disagreement over product decisions that the development team has made, and not at all what people normally think when you say "a new paper has proved GPT-4 is getting worse over time."

> what people normally think

While we're doing Keynesian beauty contests, I think that 98% of the time when people say that a product is getting worse over time, they're referring to product decisions the development team has made, and how they have been implemented.

Reply on news.ycombinator.com