QUEST: Quality-aware Semi-supervised Table Extraction for Business Documents

TL;DR

QUEST is a novel semi-supervised table extraction framework that uses a quality assessment model to improve accuracy and reduce errors in business document processing, leveraging unlabeled data effectively.

Contribution

It introduces a quality-aware pseudo-labeling approach with a new model predicting F1 scores, enhancing semi-supervised table extraction for business documents.

Findings

F1 score improved from 64% to 74% on proprietary data.

Reduces empty predictions by 45%.

Achieves 50% F1 on DocILE benchmark, up from 42%.

Abstract

Automating table extraction (TE) from business documents is critical for industrial workflows but remains challenging due to sparse annotations and error-prone multi-stage pipelines. While semi-supervised learning (SSL) can leverage unlabeled data, existing methods rely on confidence scores that poorly reflect extraction quality. We propose QUEST, a Quality-aware Semi-supervised Table extraction framework designed for business documents. QUEST introduces a novel quality assessment model that evaluates structural and contextual features of extracted tables, trained to predict F1 scores instead of relying on confidence metrics. This quality-aware approach guides pseudo-label selection during iterative SSL training, while diversity measures (DPP, Vendi score, IntDiv) mitigate confirmation bias. Experiments on a proprietary business dataset (1000 annotated + 10000 unannotated documents) show QUEST improves F1 from 64% to 74% and reduces empty predictions by 45% (from 12% to 6.5%). On the DocILE benchmark (600 annotated + 20000 unannotated documents), QUEST achieves a 50% F1 score (up from 42%) and reduces empty predictions by 19% (from 27% to 22%). The framework's interpretable quality assessments and robustness to annotation scarcity make it particularly suited for business documents, where structural consistency and data completeness are paramount.