Akari Asai

@Gera001 Hi, do you mind which eval task data do you want to learn the details of the creation process? @jonhue Thanks for the question! I guess you are intereted...

Hi! Could you give me a bit more information? Do you need custom dataset help for training data creation, or evaluation?

Hi, thank you so much for opening an issue! Sorry that the README is not really easy-to-follow. I addressed some questions briefly [here](https://github.com/AkariAsai/self-rag/issues/56) and try to improve the documentation in...

Hi! `_gs` indicates that the retrieved results are further enhanced by Google Programmable Searc in addition to the original contriever top 10 documents. We added this new results in our...

For the static inference, the retrieved passages are only retrieved at the beginning. Note that for ALCE ASQA, we use the pre-given passages provided by [the original repository](https://github.com/princeton-nlp/ALCE), which were...

Hi, sorry for my late response! I was busy with a paper release I was co-authoring in the past two weeks. Yes, if Self-RAG generates [No retrieval] at the very...

Hi @etherion-1337, thank you for your interests! yes, we are happy to release it too. I will upload the critic model as well later today. re the error: I apologize...

We started uploading the Critic model, and it should be available tomorrow. https://huggingface.co/selfrag/self_rag_critic

heh let me double check this week. We have trained multiple different variants for critic, and we might have uploaded different checkpoint -- in one of the checkpoints we have...

Hi, thank you so much for your question. I apologize for my late response. I was co-authoring another paper in the past two weeks. For the critic model, we trained...