Shepherd A Critic for Language Model Generation
Shepherd, a new model designed to critique generations from large language models (LLMs), is making waves in the AI community. Trained on community feedback and human-annotated feedback, Shepherd has demonstrated its ability to provide comprehensive critiques and identify errors in model responses. Its performance is comparable to, if not better than, existing models like Alpaca, SelFee, and ChatGPT.
You can also find more information about the paper and its authors on the provided arXiv link: https://arxiv.org/abs/2308.04592
REFERENCES
▪︎ Tianlu Wang, Ping Yu, Xiaoqing Ellen Tan, Sean O’Brien, Ramakanth Pasunuru, Jane Dwivedi-Yu, Olga Golovneva, Luke Zettlemoyer, Maryam Fazel-Zarandi, Asli Celikyilmaz. “Shepherd: A Critic for Language Model Generation.” arXiv:2308.04592 [cs.CL], submitted on 8 Aug 2023. ▪︎ Wikipedia: https://en.m.wikipedia.org/wiki/Large_language_model