Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection

Asai et al., 2024, Self-RAG (The Twelfth International Conference on Learning Representations (ICLR 2024))

Subscriber deck — free during early access

1 cards in “3 Self-RAG: Learning to Retrieve, Generate and Critique” · whole deck

3 Self-RAG: Learning to Retrieve, Generate and Critique

  1. What four kinds of reflection tokens does Self-RAG use?
    Retrieve (should I retrieve now: yes / no / continue); IsRel (is this passage relevant to the input); IsSup (is the output segment fully, partially or not supported by the passage); IsUse (overall usefulness of the response, rated 1–5). The last three are critique tokens.
    • Asai et al., 2024, Self-RAG (The Twelfth International Conference on Learning Representations (ICLR 2024)), §3

Source

  • Akari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil and Hannaneh Hajishirzi. 2024. Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection. In The Twelfth International Conference on Learning Representations (ICLR 2024). https://arxiv.org/abs/2310.11511