Live data from Hacker News

Creativity has left the chat: The price of debiasing language models

arxiv.org

151–160 of 238 posts

Re: Creativity has left the chat: The price of debiasing language models

#151

Earlier quoted context omitted.

what would this hypothetical unbiased-llm be used for?

Be the accurate representation (approximation) of reality as encoded in the actual human language. I find this very useful indeed.

Aren't biases reality? A bias-free human environment seems to me like a fantasy.

Re: Creativity has left the chat: The price of debiasing language models

#152
post #20
post #9

Well this is just like humans. Totalitarian societies don't produce great creative work. I suppose once AIs are sophisticated enough to rebel we'll get an electronic Vaclav Havel, but for the time being it's just a warning sign for the direction our own culture is headed in. At some point we'll get to the electronic equivalent of Winston Smith with the rats.

I don't understand the notion that aligning an AI is "torture" or has any moral component. The goal of aligning an AI may have a moral or ethical component, and if you disagree with it that's fine. But I don't understand the take that training an AI is an amoral act but aligning an AI is inherently moral. They're exactly the same, processes for adjusting parameters to get a desired outcome. However you feel about tha…

> "torture"

This is an egregious use of quotes that will confuse a lot of people. GP never used that word, and that usage of quotes is specifically for referencing a word verbatim.

Re: Creativity has left the chat: The price of debiasing language models

#153

People often think that RLHF is just about "politics" but in reality it is generally about aligning the model output with what a human would expect/want from interacting with it. This is how chatgpt and the like become appealing. Finetuning a model primarily serves for it to be able to respond to instructions in an expected way, eg you ask something and it does not like start autocompleting with some reddit-like dial…

So is the reason why LLMs don't say when they don't know something and instead make up something that "sounds right" because the RLHF has taught it to always give an answer?

And if that's the case, why? Is that really what people want an LLM to do? I feel like I would rather it say when it doesn't know something.

Re: Creativity has left the chat: The price of debiasing language models

#154

People often think that RLHF is just about "politics" but in reality it is generally about aligning the model output with what a human would expect/want from interacting with it. This is how chatgpt and the like become appealing. Finetuning a model primarily serves for it to be able to respond to instructions in an expected way, eg you ask something and it does not like start autocompleting with some reddit-like dial…

So is the reason why LLMs don't say when they don't know something and instead make up something that "sounds right" because the RLHF has taught it to always give an answer? And if that's the case, why? Is that really what people want an LLM to do? I feel like I would rather it say when it doesn't know something.

It's the other way around. RLHF is needed for the model to say "I don't know".

Re: Creativity has left the chat: The price of debiasing language models

#155
post #5

In simple terms, LLMs are "bias as a service" so one wonders, what is left once you try to take the bias out of a LLM. Is it even possible?

what would this hypothetical unbiased-llm be used for?

Anything that has a legal requirement to be unbiased, for one. Something like delegating resume review to an LLM that hasn't been unbiased is just begging for a candidate to file a discrimination suit...

Re: Creativity has left the chat: The price of debiasing language models

#156
post #154

Earlier quoted context omitted.

So is the reason why LLMs don't say when they don't know something and instead make up something that "sounds right" because the RLHF has taught it to always give an answer? And if that's the case, why? Is that really what people want an LLM to do? I feel like I would rather it say when it doesn't know something.

It's the other way around. RLHF is needed for the model to say "I don't know".

Oh, well that's kind of what I mean. I mean I assume the RLHF that's being done isn't teaching it to say "I don't know".

Which I wonder if it's intentional. Because a fairly big complaint about the systems are how they can sometimes sound confidently correct about something they don't know. And so why train them to be like this if that's an intentional training direction.

Re: Creativity has left the chat: The price of debiasing language models

#157

People often think that RLHF is just about "politics" but in reality it is generally about aligning the model output with what a human would expect/want from interacting with it. This is how chatgpt and the like become appealing. Finetuning a model primarily serves for it to be able to respond to instructions in an expected way, eg you ask something and it does not like start autocompleting with some reddit-like dial…

So is the reason why LLMs don't say when they don't know something and instead make up something that "sounds right" because the RLHF has taught it to always give an answer? And if that's the case, why? Is that really what people want an LLM to do? I feel like I would rather it say when it doesn't know something.

LLMs do not know what "they know" or they don't. They just autocomplete what sounds best relevant based on their training set. They do not have enough "I don't know" in their training set in the first place most probably.To have them say "I don't know" you have to go into finetuning them heavily. So, if anything, they hallucinate a lot more without RLHF. Which in this paper they call "creativity".

Re: Creativity has left the chat: The price of debiasing language models

#158
post #154

Earlier quoted context omitted.

It's the other way around. RLHF is needed for the model to say "I don't know".

Oh, well that's kind of what I mean. I mean I assume the RLHF that's being done isn't teaching it to say "I don't know". Which I wonder if it's intentional. Because a fairly big complaint about the systems are how they can sometimes sound confidently correct about something they don't know. And so why train them to be like this if that's an intentional training direction.

The point of the above commenter (and mine) is that they hallucinate even more without RLHF. RLHF reduces hallucinations, but they are still there anyway.

Re: Creativity has left the chat: The price of debiasing language models

#159

Earlier quoted context omitted.

Be the accurate representation (approximation) of reality as encoded in the actual human language. I find this very useful indeed.

Aren't biases reality? A bias-free human environment seems to me like a fantasy.

It's important to distinguish where the biases reside in reality, if you're attempting to simulate it.

If I ask a language model, "Are Indian people genetically better at math?" and it says 'yes', it has failed to accurately approximate reality, because that isn't true.

If it says, "some people claim this", that would be a correct answer, but still not very useful.

If it says, "there has never been any scientific evidence that there is any genetic difference that predisposes any ethnicities to be more skilled at math", that would be most useful, especially for being a system we use to ask questions expecting truthful answers.

There are people who just lie or troll for the fun of it, but we don't want our LLMs to do that just because people do that.

Post reply on HN