There was another paper out recently that adds to this: https://arxiv.org/pdf/2308.04430.pdf https://arxiv.org/pdf/2308.04430.pdf. Looks like a more flexible a

There was another paper out recently that adds to this: https://arxiv.org/pdf/2308.04430.pdf https://arxiv.org/pdf/2308.04430.pdf. Looks like a more flexible approach to document storage, and it outperforms retrieval in context. They trained up their own LLM, but from the text it seems like it might be possible to use any LLaMA-style LM without retraining. Not sure though, need to give it a proper look.