This includes the original implementation of SELF-RAG: Learning to Retrieve, Generate and Critique through self-reflection by Akari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil, and Hannaneh Hajishirzi.
It seems that your team is training two models for self-RAG. I am curious if the model your team has uploaded to Huggingface, named selfrag/selfrag_llama2_7b, is a Critic or Generator model.？
It seems that your team is training two models for self-RAG. I am curious if the model your team has uploaded to Huggingface, named selfrag/selfrag_llama2_7b, is a Critic or Generator model.？