Paper Overview
- Field: Computer Vision (CV)
- Authors: Xiazhen Wu, Wansong Qin, Yangbin Zheng, Liangda Fang, Zhan Li, Xiujie Huang, Liushen Zhou, Quanlong Guan
- Published: 2026-10-08
- arXiv: 2610.12363
- HANS poses significant challenges to existing document parsing and handwriting recognition methods.
- NA-GOT achieves notable improvements in both accuracy and stability for answer sheet process recognition.
- The dataset will be made publicly available upon publication.
Summary
Intelligent grading and automated scoring technologies constitute critical infrastructure for smart education. However, existing document parsing and handwriting recognition benchmarks are predominantly designed for well-structured printed documents or isolated mathematical expressions, lacking datasets that capture the complex characteristics inherent to student answer sheets, including multi-line derivation processes, heterogeneous mixtures of text and mathematical formulae, and noise artifacts such as strikethroughs.
To address this gap, the authors introduce HANS, the first dataset explicitly constructed for real-world educational scenarios. It encompasses mathematical expressions, natural language text, hand-drawn tables, and diverse noise patterns including corrections and deletions, accompanied by fine-grained annotations, laying a reliable foundation for research on robust recognition.
Building on HANS, they propose NA-GOT, an end-to-end framework that achieves two-level noise suppression through a lightweight feature-level noise suppression module and a noise-aware attention mechanism incorporated into the decoding stage.
Key Findings
*Originally posted on zhichai.net; auto-collected on 2026-10-12.*