Awesome RL with Human Feedback / 2024 / arXiv

Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback

作者：Jiaming Ji, Jiayi Zhou, Hantao Lou, Boyuan Chen, Donghai Hong, Xuyao Wang, Wenqi Chen, Kaile Wang, Rui Pan, Jiahao Li, Mohan Wang, Josef Dai, Tianyi Qiu, Hua Xu, Dong Li, Weipeng Chen, Jun Song, Bo Zheng, Yaodong Yang · 单位：Institute for AI, Peking University, Beijing Academy of Artificial Intelligence (BAAI), Huawei Noah’s Ark LAB, Taobao & Tmall Group of Alibaba · 来源：arXiv · 发布日期：2024

对齐目标设计评测/基准数据筛选多模态反馈人类反馈设计
新标签打开 AlphaXiv新标签打开 Code
Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback figure
AlphaXiv 中文概览（可滚动查看）