RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback · HackerPapers discussions

RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback — mentioned by esseph

Tue, 08 Sep 2026 02:15:01 GMT

> Im assuming the people who _train_ AI [ LLMs ] actually need to separate the two, and avoid the feedback loop of training the next LLM on the output of the previous LLM. Check out Reinforcement Learning from AI Feedback (RLAIF). Then skim some of this maybe: https://arxiv.org/abs/2309.00267

Paper and discussion history. A mention is not an endorsement.

RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback — mentioned by astrange

Fri, 24 May 2024 02:51:48 GMT

The previous model does. https://arxiv.org/abs/2309.00267 That's during fine-tuning though. I know I read one about applying it during pretraining, maybe this one? https://arxiv.org/abs/2302.08582

Paper and discussion history. A mention is not an endorsement.

RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback — HN submission

Mon, 04 Sep 2023 07:31:39 GMT

Rlaif: Scaling Reinforcement Learning from Human Feedback with AI Feedback

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.