Work

Retrieval-Conditional Parsing Score (RCPS): Choosing Document Parsers by Retrieval, Not by Appearance

Year: 2026
Type: conference-paper
Abstract: Document RAG retrieves from parser outputs, yet parsers are commonly selected by intrinsic measures of transcription fidelity and boundary quality rather than by retrieval performance.We introduce the Retrieval-Conditional Parsing Score (RCPS), a trainingfree protocol that ranks parser-chunker combinations on a fixed held-out Q-A retrieval probe.RCPS averages mean reciprocal rank (MRR) across specified retrievers and retrieval depths, using the same format-normalised exact-span relevance rule for every candidate.On KoGovDoc-RAG, table-enabled MinerUon has a higher Boundary Clarity score than our production parser, Prod (0.713 versus 0.610).Yet its Hit@1 is 42.6 percentage points lower (0.123 versus 0.549).Across five parsers evaluated on all 294 pages, RCPS ranges from 0.137 to 0.584.With Prod fixed, the four chunkers score between 0.535 and 0.593.A retriever-free coverage diagnostic finds no normalised exact match for 20.2% of reference spans in Prod output.No tested chunker splits more than 2.3%.On sourcealigned OHR-Bench data, semantic noise lowers retrieval while Boundary Clarity remains stable or changes non-monotonically.In a secondary parser-training study, two direct preference optimization checkpoints improve Hit@5 by only 0.95 and 1.15 percentage points over Prod on an audited compatibility subset; a matched edit-distance control performs similarly.The workflow therefore selects with RCPS, diagnoses with coverage, and trains only if needed.We release the 663-Q-A KoGovDoc-RAG probe and an RCPS reference implementation.Evaluation frame Candidate pipelines RCPS evaluation Select, diagnose + act
Language: English
Open Access status: hybrid
OpenAlex ID: W7224323931
FWCI:
Cites: 24
Cited by:
Related to: 10