Papers for

document processing teams

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Vision language models trade off accuracy cost and governance in document extraction

Beyond Accuracy: Robustness, Cost, and Governance Trade-offs for Vision-Language Models in Templated Document Extraction

Abstract: Vision-language models (VLMs) are increasingly used to extract structured fields from business documents, yet most evaluations report accuracy on clean benchmarks and offer little guidance to practitioners choosing an approach for a given task complexity. We address this gap with a measurement-grounded study and an open-source release. Across eleven systems (three commercial, two reasoning, five open-source VLMs in pretrained and fine-tuned form, and a non-LLM OCR->regex floor) scored on a 750-document held-out pool of synthetic checks, fine-tuning on 3K samples lifts the best open-source VLMs above F1 0.98-above every zero-shot commercial system on this task-while GPT-5 leads the commercial pool on F1 and Claude Sonnet 4.5 collapses on Date. To turn these measurements into actionable choices, we introduce a practitioner-oriented selection framework that maps a task profile (quality, latency, governance, volume) to a recommended approach via filtering and total-cost minimization, illustrated on a hypothetical mid-volume document-extraction scenario.

Mon 14 SeptArtificial Intelligence
The gist
Extracting important information from business documents often uses computer programs that understand both images and text. This study compared eleven different systems on a set of 750 sample documents and found that some open-source models, when trained with enough examples, performed very well — even better than several commercial systems. The researchers also noticed that the performance can vary a lot depending on the specific task, like extracting dates. They created a practical guide to help people choose the best approach based on quality, speed, rules, and cost.
Open → 2609.15706v1

Bounding box guidance improves table structure recognition accuracy

Enhancing Table Structure Recognition via Bounding Box Guidance

Abstract: Table Structure Recognition (TSR) aims to extract the bounding boxes of cells and table structure (e.g., HTML) from table images. Although current approaches have made significant progress, the latest image-to-sequence methods overlook the explicit utilization of the bounding box information when predicting HTML sequences, leading to error predictions in complex scenes. In this paper, we introduce a novel framework BGTR (Bounding Box-Guided Table Recognizer). To more effectively utilize bounding box information, we first predict the bounding boxes of cells and then use this information to guide the generation of HTML sequences. While utilizing bounding box information can enhance the accuracy of HTML sequences, for natural scene tables, the data volume is too small to allow for sufficient training of bbox-guided HTML generation. In response, we adopt a progressive training method for natural scene tables and introduce SNSTab, a synthetically generated natural scene table dataset. Our experiments on five benchmark datasets demonstrate SOTA performance.

Tue 8 SeptComputer Vision and Pattern Recognition
The gist
Recognizing tables in images means figuring out where each cell is and how they fit together. The authors noticed that many current methods miss using the exact locations of table cells when turning an image into a code-like description. They propose a new method that first finds the locations of cells and then uses this to better predict the table’s structure. To train and test this method especially on photos of tables in natural scenes, they created a synthetic dataset. Their approach showed improved accuracy across several tests.
Open → 2609.08705v1