arXiv:2608. 06607v1 Announce Type: cross Abstract: Most document-extraction systems use a single model for all documents.
arXiv:2609.01575v1 Announce Type: new Abstract: Extracting structured fields from hundreds of millions of documents annually remains costly in regulated industries: bespoke OCR cascades cover only a...
The paper evaluates eleven vision‑language models (VLMs) for extracting structured fields from business documents, focusing on robustness, cost, and governance rather than just accuracy. Using a held‑out set of 750 synthetic checks, the study finds that fine‑tuning open‑source VLMs on 3,000 samples yields an F1 score above 0.98, surpassing all zero‑shot commercial systems, while GPT‑5 tops the commercial group and Claude Sonnet 4.5 fails on date extraction. The authors also present a practitioner‑oriented selection framework that maps task profiles—such as quality, latency, governance, and volume—to recommended approaches via filtering and total‑cost minimization, demonstrated on a mid‑volume document‑extraction scenario.
The study evaluates machine‑generated text detectors on scientific literature, finding that many detectors perform near perfectly on certain benchmarks due to superficial formatting cues rather than genuine linguistic differences. A logistic regression using 27 superficial features achieved macro‑F1 scores between 0.833 and 0.997, outperforming other detectors on arXiv subsets. The most powerful cue was the PDF extraction artifact—human texts preserve line breaks while generated texts appear as a single block—allowing classification at 0.996 AUC. By introducing format counterfactuals and normalizing formatting, the authors demonstrate that re‑wrapping machine‑written abstracts can flip 20–27% of detector decisions, reducing the generalization gap and improving out‑of‑domain performance.
The paper investigates the gap between retrieval and reading in document vision‑language models, showing that even when the correct page is retrieved, the model often fails to use the text on that page. By comparing answers derived from page images alone versus images plus extracted OCR text, the authors demonstrate that adding OCR text can improve strict accuracy by 13 to 16 points on their FoveDoc‑Bench benchmark. The study also reveals that OCR benefits textual evidence but not charts or figures, and that its advantage diminishes as retrieval quality worsens.
arXiv:2609.22628v1 Announce Type: new Abstract: Every document QA system begins with a choice that is rarely studied on its own: whether to feed the model page images, extracted text, or both. We iso...
arXiv:2607. 29677v1 Announce Type: new Abstract: Enterprise workflows increasingly rely on agents for \emph{schema-guided extraction}: given a document and a user-defined schema, the agent faithfully follows the schema to produce the correct output with source evidence as grounding metadata.
The paper evaluates open-source OCR, LLM, and VLM systems on a high‑risk public sector task: extracting structured data from student application documents. Results show that VLMs generally outperform OCR+LLM pipelines, yet only 4 of 35 configurations achieve F1 scores above 0.5, with most combinations scoring below 0.25. Model size and input quality, especially preserving OCR structure, are critical factors influencing performance.
The paper introduces a scalable product‑linking system that uses a retrieve‑then‑match cascade. First, a lightweight text cross‑encoder auto‑resolves the majority of merchant‑catalog product pairs with high precision, while an agentic multimodal vision‑language model handles the remaining ambiguous cases by inspecting images and performing web searches. This approach balances computational cost and accuracy, improving overall link coverage from 68% to 77% without requiring fine‑tuning of the agent.
The paper audits the RVL-CDIP document classification benchmark, uncovering that about 10.5% of its 400,000 documents have incorrect labels and that 35.2% of the test set contains duplicates or near-duplicates in the training set. It shows that identification codes on pages enable 69% classification accuracy, and removing them drops performance by 11 points; OCR engine choice can shift accuracy by 4.1 to 8.3 points. After correcting defects, 13 re‑trained models improve by 0.9 to 2.9 points, with LayoutLMv3 achieving 96.5% accuracy, and the authors release a cleaned corpus, annotations, and out‑of‑distribution data.
SHELF is a Python system that creates synthetic, controlled benchmark data for evaluating large language models on bibliographic tasks such as classification, clustering, retrieval, pair classification, and instruction retrieval. It generates 62,899 model-written documents based on Library of Congress vocabularies and compares methods like TF, TF‑IDF, BM25, popular encoders, and zero‑shot decoders, reporting performance metrics such as 0.8887 for subject classification and 0.2605 for genre‑form classification. The tool also allows independent variation of bibliographic facets and can produce unseen documents beyond a model’s training cutoff, with results indicating that model rankings transfer more reliably than absolute scores when compared to other benchmarks.
arXiv:2603. 26815v3 Announce Type: replace-cross Abstract: Retrieval-Augmented Generation (RAG) systems for financial document QA typically follow a chunk-based paradigm: documents are split into fragments, embedded, and retrieved by similarity.
arXiv:2609.13279v1 Announce Type: new Abstract: Fashion attribute extraction is evaluated inconsistently: results are reported as single aggregate numbers across image types that pose different probl...