RLHF decreases AI Model Creativity
Andrej Karpathy says RLHF is barely better than RL and only in situations that are open-ended and difficult for humans to do from scratch (like “write a poem”).
sometimes I’m a bit surprised RLHF works for LLMs at all. The RM we train for LLMs is just a vibe check in the exact same way. It gives high scores to the kinds of assistant responses that human raters statistically seem to like. It’s not the “actual” objective of correctly solving problems, it’s a proxy objective of what looks good to humans. Second, you can’t even run RLHF for too long because your model quickly learns to respond in ways that game the reward model. These predictions can look really weird, e.g. you’ll see that your LLM Assistant starts to respond with something non-sensical like “The the the the the the” to many prompts. Which looks ridiculous to you but then you look at the RM vibe check and see that for some reason the RM thinks these look excellent. Your LLM found an adversarial example. It’s out of domain w.r.t. the RM’s training data, in an undefined territory. Yes you can mitigate this by repeatedly adding these specific examples into the training set, but you’ll find other adversarial examples next time around. For this reason, you can’t even run RLHF for too many steps of optimization. You do a few hundred/thousand steps and then you have to call it because your optimization will start to game the RM. This is not RL like AlphaGo was.
And yet, RLHF is a net helpful step of building an LLM Assistant. I think there’s a few subtle reasons but my favorite one to point to is that through it, the LLM Assistant benefits from the generator-discriminator gap. That is, for many problem types, it is a significantly easier task for a human labeler to select the best of few candidate answers, instead of writing the ideal answer from scratch. A good example is a prompt like “Generate a poem about paperclips” or something like that. An average human labeler will struggle to write a good poem from scratch as an SFT example, but they could select a good looking poem given a few candidates. So RLHF is a kind of way to benefit from this gap of “easiness” of human supervision. There’s a few other reasons, e.g. RLHF is also helpful in mitigating hallucinations because if the RM is a strong enough model to catch the LLM making stuff up during training, it can learn to penalize this with a low reward, teaching the model an aversion to risking factual knowledge when it’s not sure. But a satisfying treatment of hallucinations and their mitigations is a whole different post so I digress. All to say that RLHF is net useful, but it’s not RL.
Rohan Paul summarizes a new paper
While alignment reduces toxic and biased content, it also seems to limit their creativity, defined as the ability to generate outputs with high syntactic and semantic diversity.
https://arxiv.org/abs/2406.05587
Creativity Has Left the Chat: The Price of Debiasing Language Models