Symbol-Aware Reasoning with Masked Discrete Diffusion for Handwritten Mathematical Expression Recognition

TL;DR

This paper introduces a discrete diffusion approach for handwritten mathematical expression recognition, focusing on iterative symbolic refinement to improve structural accuracy and robustness over traditional autoregressive models.

Contribution

It proposes a novel diffusion-based framework that reformulates HMER as a multi-step refinement process, enhancing structural consistency and handling handwriting diversity.

Findings

Achieves 5.56% CER and 60.42% EM on MathWriting benchmark.

Outperforms strong Transformer and commercial baselines.

Demonstrates consistent improvements on CROHME datasets.

Abstract

Handwritten Mathematical Expression Recognition (HMER) requires reasoning over diverse symbols and 2D structural layouts, yet autoregressive models struggle with exposure bias and syntactic inconsistency. We present a discrete diffusion framework that reformulates HMER as iterative symbolic refinement instead of sequential generation. Through multi-step remasking, the proposal progressively refines both symbols and structural relations, removing causal dependencies and improving structural consistency. A symbol-aware tokenization and Random-Masking Mutual Learning further enhance syntactic alignment and robustness to handwriting diversity. On the MathWriting benchmark, the proposal achieves 5.56\% CER and 60.42\% EM, outperforming strong Transformer and commercial baselines. Consistent gains on CROHME 2014--2023 demonstrate that discrete diffusion provides a new paradigm for structure-aware visual recognition beyond generative modeling.