# The Return of Structural Handwritten Mathematical Expression Recognition

arXiv: https://arxiv.org/abs/2508.19773 · HTML: https://mlsanity.com/paper/2508.19773/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 24 references that are in the map.
- First posted on arXiv 27 Aug 2025.

## Builds on (most cited first, 5 of 5)

- [Handwritten Mathematical Expression Recognition with Bidirectionally Trained Transformer](https://mlsanity.com/paper/2105.02412/) (2021)
- [CoMER: Modeling Coverage for Transformer-based Handwritten Mathematical Expression Recognition](https://mlsanity.com/paper/2207.04410/) (2022)
- [MathWriting: A Dataset For Handwritten Mathematical Expression Recognition](https://mlsanity.com/paper/2404.10690/) (2024)
- [Edge-Featured Graph Attention Network](https://mlsanity.com/paper/2101.07671/) (2021)
- [Local and Global Graph Modeling with Edge-weighted Graph Attention Network for Handwritten Mathematical Expression Recognition](https://mlsanity.com/paper/2410.18555/) (2024)

## Built on by (most cited first, 1 of 1)

- [Cross-Domain Few-Shot Writer Adaptation for Real-World Handwritten Mathematical Expression Recognition](https://mlsanity.com/paper/2609.32513/) (2026)

## Abstract (arXiv)

Handwritten Mathematical Expression Recognition is foundational for educational technologies, enabling applications like digital note-taking and automated grading. While modern encoder-decoder architectures with large language models excel at LaTeX generation, they lack explicit symbol-to-trace alignment, a critical limitation for error analysis, interpretability, and spatially aware interactive applications requiring selective content updates. This paper introduces a structural recognition approach with two innovations: 1 an automatic annotation system that uses a neural network to map LaTeX equations to raw traces, automatically generating annotations for symbol segmentation, classification, and spatial relations, and 2 a modular structural recognition system that independently optimizes segmentation, classification, and relation prediction. By leveraging a dataset enriched with structural annotations from our auto-labeling system, the proposed recognition system combines graph-based trace sorting, a hybrid convolutional-recurrent network, and transformer-based correction to achieve competitive performance on the CROHME-2023 benchmark. Crucially, our structural recognition system generates a complete graph structure that directly links handwritten traces to predicted symbols, enabling transparent error analysis and interpretable outputs.