Research map / Large language models / Agents, code and math
Complex Mathematical Expression Recognition: Benchmark, Large-Scale Dataset and Strong Baseline
Weikang Bai, Yongkun Du, Yuchen Su, Yazhen Xie, Zhineng Chen
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 |
|---|---|---|---|
| train a specialized small model on CMER-3M with SML and custom tokenizer instead of general MLLMs | CMERNet (125M) · CMER-Bench Complex | BLEU: 0.3089 -> 0.5568 | Improved |
| CMERNet vs UniMERNet on complex expressions | CMERNet · CMER-Bench Complex | BLEU: 0.4093 -> 0.5568 | Improved |
| CMERNet vs baselines under CDM metric | CMERNet · CMER-Bench Moderate | CDM: 0.886 -> 0.845 | Hurt |
| CMERNet vs Gemini-2.5-pro under CDM on complex tier | CMERNet · CMER-Bench Complex | CDM: 0.856 -> 0.690 | Hurt |
| general-purpose MLLMs on complex vs easy expressions | Gemini-2.5-pro · CMER-Bench | BLEU: 0.455 -> 0.266 | Hurt |
| CMERNet vs UniMERNet on complex edit distance | CMERNet · CMER-Bench Complex | avg edit distance: 483.34 -> 474.18 | No clear effect |
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.
- Dolphin — DOLPHIN: A Programmable Framework for Scalable Neurosymbolic Learning (ICML 2025) · better on 3 of 4 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 0.87)
- TAMER: Tree-Aware Transformer for Handwritten Mathematical Expression Recognition (AAAI 2025 · similarity 0.86)
- MathPile: A Billion-Token-Scale Pretraining Corpus for Math (NeurIPS 2024 · similarity 0.84)
- MathSpeech: Leveraging Small LMs for Accurate Conversion in Mathematical Speech-to-Formula (AAAI 2025 · similarity 0.84)
- SSAN: A Symbol Spatial-Aware Network for Handwritten Mathematical Expression Recognition (AAAI 2025 · similarity 0.84)
- MathCoder2: Better Math Reasoning from Continued Pretraining on Model-translated Mathematical Code (ICLR 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.
