W-RAC decouples extraction from semantic planning via structured units and LLM grouping to match traditional retrieval performance at roughly 10x lower LLM token cost.
All text, hyperlinks, links, formatting, images, image links, and elements must remain exactly as in the original markdown and present in the output chunks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Web Retrieval-Aware Chunking (W-RAC) for Efficient and Cost-Effective Retrieval-Augmented Generation Systems
W-RAC decouples extraction from semantic planning via structured units and LLM grouping to match traditional retrieval performance at roughly 10x lower LLM token cost.