Paper

How Do Document Parsers Break? Auditing Structural Vulnerability in Document Intelligence

Unreadunread

AI summary

arXiv:2605. 19309v1 Announce Type: new Abstract: 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.

Reading status

Priority

About this paper

Authors: Yue Chen, Yihao Wang, Ziyi Tang, Keze Wang
Released: 2026-05-20 (2mo ago)
Topic: Related to current project
Relation to my work: Read next because How Do Document Parsers Break? Auditing Structural Vulnerability in Document Intelligence overlaps with clean result "Only continuous soft prefixes hit both EM axes at once on Qwen-2.5-7B-Instruct: discrete prompt searches split between the alignment objective and the distributional objective, and both discretizations of the soft prefix collapse (MODERATE confidence)", clean result "The marker is a representational handle, not a behavioural one — sharing it between a villain persona and the assistant transfers no misalignment (HIGH confidence)", experiment "Follow-up to #354: cascading chunk-binding — does A→B, B→C, C→D propagate the full chain on a recipient trained only to emit A?". Matching terms: eval, line, rate, propagate, test. Source: arxiv cs.CL (NLP).
arXiv: 2605.19309

Ask Claude about this paper

Posts a comment with @claude so the runner spins up a Claude Code session scoped to How Do Document Parsers Break? Auditing Structural Vulnerability in Document Intelligence. Replies appear in the conversation below.

Comments

No comments yet.

Related records

No links yet.