Towards Pareto-Efficient RLHF: Paying Attention to a Few High-Reward Samples with Reward Dropout

Towards Pareto-Efficient RLHF: Paying Attention to a Few High-Reward Samples with Reward Dropout Changhun Lee author Chiehyeon Lim author 2024-11 text Findings of the Association for Computational Linguistics: EMNLP 2024 Yaser Al-Onaizan editor Mohit Bansal editor Yun-Nung Chen editor Association for Computational Linguistics Miami, Florida, USA conference publication lee-lim-2024-towards 10.18653/v1/2024.findings-emnlp.489 https://aclanthology.org/2024.findings-emnlp.489/ 2024-11 8335 8349