retrieval · retrieval-augmented-generation TEC-0020

Ground answers in retrieved documents

Retrieve relevant documents and generate from them, with the model deciding when retrieval is needed.

Retrieve passages for the query and generate an answer that cites them. Self-RAG extends this with reflection tokens so the model decides when to retrieve and checks whether each passage supports what it wrote.

Addresses: Stating false facts confidently, Checking claims against evidence

Contexts: Retrieval-augmented QA, Chat assistant

Does it work?

supported so far2 supporting · 0 contesting sources · last moved 2026-09-04

Efficacy claims — what this technique actually moves, under which conditions, and whether that has been contested.

Needs: documents or an index to retrieve from

Fails when:

Retrieved content carries planted instructions, or the wrong account is better represented among conflicting sources. Whatever lands in context is treated as reliable. (helps, narrowly)

1 more condition from claims reviewed by AI, listed with the claims below.

mechanismreplicated
Grounding answers in retrieved documents reduces unsupported generation but transfers the trust problem to the retrieval channel: whatever lands in context is treated as reliable, which shows up both as following instructions planted in retrieved content and as deferring to whichever account is better represented when sources disagree.
- supports Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection (2023)
- supports Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflicts (2023)

Code

https://github.com/AkariAsai/self-rag — Self-RAG training and inference code.verified 2026-09-02

Sources

Either way, bring evidence: a link, or what you saw.

Whether a technique works is a claim. What you file here becomes a claim about this technique under the first capability it addresses, with your evidence as its source, and moves the standing above the same way a paper would.