Active Retrieval Augmented LLMs (FLARE)
Free while signed in. Answers cite the passages they came from.
First page
The curator’s take
Key pointsActively decides when and what to retrieve during generation.
01
Dynamic retrieval: Retrieves only when the model's next-token confidence drops - not at fixed intervals.
02
Anticipated content retrieval: Retrieves based on what the model is about to generate, not just the current context.
03
Long-form knowledge-intensive tasks: Demonstrates superior or competitive performance on long-form knowledge-intensive generation tasks.
04
Adaptive RAG: Established a research direction on adaptive/active retrieval that matured in 2024 with tools like Self-RAG and RankRAG.
Every Monday
Get next week’s papers.
The same picks and the same summaries, in your inbox. Free, and 176 issues deep.
Subscribe on Substack