Research map / Architectures and efficiency / Attention and transformers
PosFormer: Recognizing Complex Handwritten Mathematical Expression with Position Forest Transformer
Tongkun Guan, Chengyu Lin, Wei Shen, Xiaokang Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Position Forest auxiliary position recognition task (ablation) | CoMER baseline · CROHME 2014 | ExpRate: 59.33 -> 62.13 | Improved |
| add Position Forest auxiliary position recognition task (ablation) | CoMER baseline · CROHME 2019 | ExpRate: 62.97 -> 63.80 | Improved |
| add Implicit Attention Correction on top of Position Forest (ablation) | PosFormer · CROHME 2014 | ExpRate: 62.13 -> 62.64 | Improved |
| add Implicit Attention Correction on top of Position Forest (ablation) | PosFormer · CROHME 2016 | ExpRate: 61.03 -> 61.20 | Improved |
| add Position Forest to RNN-based DWAP (ablation) | DWAP · CROHME 2014 | ExpRate: 50.10 -> 57.10 | Improved |
| add Position Forest to ABM (ablation) | ABM · CROHME 2019 | ExpRate: 53.96 -> 54.30 | Improved |
| add Position Forest to CAN (ablation) | CAN · CROHME 2019 | ExpRate: 52.96 -> 57.13 | Improved |
| use Position Forest versus RobustScanner position enhancement on DWAP (ablation) | DWAP · Average over CROHME 14/16/19 | ExpRate: 53.18 -> 56.87 | Improved |
| use Position Forest versus the SAN tree decoder as auxiliary on DWAP (ablation) | DWAP · CROHME 2014 | ExpRate: 50.41 -> 57.10 | Improved |
| PosFormer versus CoMER as nesting depth grows (N1, N2, N3) | PosFormer · MNE test set | ExpRate: +0.86 (N1); +1.65 (N2); +10.04 (N3) | Improved |
| Position Forest is used only in training and removed at inference | PosFormer · Inference speed vs CoMER | FPS: 6.30 -> 6.28 | No clear effect |
Later papers that compared with this method
1 later accepted paper reports PosFormer as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- UniMERNet: A Universal Network for Real-World Mathematical Expression Recognition (CVPR 2026 · better on 2 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CoMER — CoMER: Modeling Coverage for Transformer-Based Handwritten Mathematical Expression Recognition (ECCV 2022) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Uni-MuMER: Unified Multi-Task Fine-Tuning of Vision-Language Model for Handwritten Mathematical Expression Recognition (NeurIPS 2025 · similarity 0.91)
- SSAN: A Symbol Spatial-Aware Network for Handwritten Mathematical Expression Recognition (AAAI 2025 · similarity 0.90)
- TAMER: Tree-Aware Transformer for Handwritten Mathematical Expression Recognition (AAAI 2025 · similarity 0.90)
- Generating Handwritten Mathematical Expressions From Symbol Graphs: An End-to-End Pipeline (CVPR 2024 · similarity 0.88)
- Image Over Text: Transforming Formula Recognition Evaluation with Character Detection Matching (CVPR 2025 · similarity 0.82)
- On the Generalization of Handwritten Text Recognition Models (CVPR 2025 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
