Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chen, Yue, Wang, Yihao, Tang, Ziyi, Zheng, Yongsen, Wang, Keze
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.19309
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910260186316800
author Chen, Yue
Wang, Yihao
Tang, Ziyi
Zheng, Yongsen
Wang, Keze
author_facet Chen, Yue
Wang, Yihao
Tang, Ziyi
Zheng, Yongsen
Wang, Keze
contents Document Layout Analysis (DLA) pipelines provide structured page representations for retrieval-augmented generation, long-document question answering, and other document intelligence systems, yet their robustness evaluation remains largely area-centric. We identify this Footprint Bias and propose ProSA, a lightweight output-level auditing framework that decouples controlled probing, policy-driven targeting, and structure-aware diagnosis. ProSA combines Block-level Structural Loss Rate (B-SLR), granularity-aware exposure descriptors, and pathway attribution to analyze where structural identity is lost, at what exposure granularity failures emerge, and how failures propagate. Across MinerU and PP-StructureV3 on 1,000 pages, affected area weakly tracks perturbation-induced OCR instability (R^2=0.384/0.110), whereas B-SLR aligns much more closely with it (R^2=0.727/0.916). Exposure descriptors further separate occlusion- and topology-dominant pathways, while matched-footprint structural probes cause much larger downstream QA/retrieval degradation compared to area-matched erasure. These results shift DLA robustness evaluation from footprint-based stress testing toward structure-aware vulnerability auditing.
format Preprint
id arxiv_https___arxiv_org_abs_2605_19309
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle How Do Document Parsers Break? Auditing Structural Vulnerability in Document Intelligence
Chen, Yue
Wang, Yihao
Tang, Ziyi
Zheng, Yongsen
Wang, Keze
Computation and Language
Document Layout Analysis (DLA) pipelines provide structured page representations for retrieval-augmented generation, long-document question answering, and other document intelligence systems, yet their robustness evaluation remains largely area-centric. We identify this Footprint Bias and propose ProSA, a lightweight output-level auditing framework that decouples controlled probing, policy-driven targeting, and structure-aware diagnosis. ProSA combines Block-level Structural Loss Rate (B-SLR), granularity-aware exposure descriptors, and pathway attribution to analyze where structural identity is lost, at what exposure granularity failures emerge, and how failures propagate. Across MinerU and PP-StructureV3 on 1,000 pages, affected area weakly tracks perturbation-induced OCR instability (R^2=0.384/0.110), whereas B-SLR aligns much more closely with it (R^2=0.727/0.916). Exposure descriptors further separate occlusion- and topology-dominant pathways, while matched-footprint structural probes cause much larger downstream QA/retrieval degradation compared to area-matched erasure. These results shift DLA robustness evaluation from footprint-based stress testing toward structure-aware vulnerability auditing.
title How Do Document Parsers Break? Auditing Structural Vulnerability in Document Intelligence
topic Computation and Language
url https://arxiv.org/abs/2605.19309