Prototypical Reward Network for Data-Efficient RLHF

Jan 1, 2024·
Yiqiao Jin
Yiqiao Jin
,
Xiting Wang
,
Yaru Hao
,
Xing Xie
· 1 min read
Abstract
We propose a prototypical reward network that enables data-efficient reinforcement learning from human feedback (RLHF) for large language models.
Type
Publication
Annual Meeting of the Association for Computational Linguistics (ACL) 2024

Abstract

We propose a prototypical reward network that enables data-efficient reinforcement learning from human feedback (RLHF) for large language models.

Keywords

Reinforcement Learning, Human Feedback, Large Language Models, Data Efficiency

Reinforcement Learning Human Feedback Large Language Models Data Efficiency
Yiqiao Jin
Authors
Yiqiao Jin
CS PhD Student
My research interests include large language models (LLMs), multi-agent systems (MASs), and multimodal models.