Research map / Image generation / Text-to-image diffusion
From Pixel to Precision: Enhancing Handwritten Mathematical Expression Recognition with Image-Level Reward
Ze Liu, Kai Zhang, Xianquan Wang, Shuochen Liu, Jiaxian Yan, Yupeng Han, Qi Liu
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 |
|---|---|---|---|
| fine-tune with IMPO (GRPO with IMS image reward) | PosFormer · CROHME 2019 | ExpRate (%): 64.97 -> 66.34 | Improved |
| fine-tune with IMPO | PosFormer · HME100K | ExpRate (%): 69.51 -> 70.67 | Improved |
| fine-tune with IMPO | PosFormer · M2E | ExpRate (%): 58.33 -> 59.56 | Improved |
| fine-tune with IMPO | ABM · CROHME 2014 | ExpRate (%): 56.85 -> 57.98 | Improved |
| fine-tune with IMPO | CoMER · CROHME 2014 | ExpRate (%): 59.33 -> 60.72 | Improved |
| use IMS reward instead of SSIM under GRPO (ablation) | CoMER · CROHME 2014 | ExpRate (%): 60.11 -> 60.72 | Improved |
| use GRPO instead of REINFORCE with baseline for IMS (ablation) | CoMER · CROHME 2014 | ExpRate (%): 59.77 -> 60.72 | Improved |
| column-wise IMS instead of horizontal division (ablation) | CoMER · CROHME 2014 | ExpRate (%): 59.68 -> 60.72 | Improved |
| remove blank columns in IMS (ablation) | CoMER · CROHME 2014 | ExpRate (%): 60.36 -> 60.72 | Improved |
| use IMS reward instead of BLEU/ROUGE text rewards (ablation) | CoMER · CROHME 2014, long formulas | ExpRate (%): 25.48 -> 27.65 | Improved |
| use text-level BLEU or ROUGE reward (ablation) | CoMER · CROHME 2014 formulas 40+ | ExpRate (%): 25.48 -> 25.87 | No clear effect |
| use IMS instead of SSIM for rendering robustness across DPI (ablation) | IMS · CROHME 2014 DPI 150-300 | coefficient of variation (%): 1.21 -> 0.62 | Improved |
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 0.89)
- Generating Handwritten Mathematical Expressions From Symbol Graphs: An End-to-End Pipeline (CVPR 2024 · similarity 0.87)
- TAMER: Tree-Aware Transformer for Handwritten Mathematical Expression Recognition (AAAI 2025 · similarity 0.87)
- Image Over Text: Transforming Formula Recognition Evaluation with Character Detection Matching (CVPR 2025 · similarity 0.84)
- Table2LaTeX-RL: High-Fidelity LaTeX Code Generation from Table Images via Reinforced Multimodal Language Models (NeurIPS 2025 · similarity 0.83)
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.
