MacRAG retrieves small slices of compressed summaries, scales up candidate chunks, and merges neighboring chunks, achieving higher average F1 than RAPTOR and LongRAG on three LongBench multi-hop QA datasets.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MacRAG: Compress, Slice, and Scale-up for Multi-Scale Adaptive Context RAG
MacRAG retrieves small slices of compressed summaries, scales up candidate chunks, and merges neighboring chunks, achieving higher average F1 than RAPTOR and LongRAG on three LongBench multi-hop QA datasets.