Research map / Probabilistic and scientific modeling / Differential equations and PDEs
Deep Learning For Symbolic Mathematics
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Increase beam width from 1 to 50 | Transformer seq2seq · ODE order 2, held-out test set | accuracy: 43.014 -> 81.166 | Helped |
| Increase beam width from 1 to 50 | Transformer seq2seq · ODE order 1, held-out test set | accuracy: 77.582 -> 93.973 | Helped |
| Increase beam width from 10 to 50 | Transformer seq2seq · Integration (BWD), held-out test set | accuracy: 99.4 -> 99.7 | Helped |
| Train with BWD + IBP data instead of BWD data alone | Integration model · Integration (FWD) test set | accuracy: 27.5228 -> 56.1168 | Helped |
| Add FWD examples to BWD + IBP training data | Integration model · Integration (FWD) test set | accuracy: 56.1168 -> 94.2609 | Helped |
| Train only on FWD data rather than mixed generators | Integration model · Integration (BWD) test set | accuracy: 99.6714 -> 17.1979 | Hurt |
| Train only on BWD data rather than mixed generators | Integration model · Integration (FWD) test set | accuracy: 94.2609 -> 27.5228 | Hurt |
| Use larger model sizes (ablation) | Transformer seq2seq · Integration and ODE tasks | accuracy | No clear effect |
Closest papers by meaning
- A Neural-Guided Dynamic Symbolic Network for Exploring Mathematical Expressions from Data (ICML 2024 · similarity 0.88)
- Neuro-Symbolic Data Generation for Math Reasoning (NeurIPS 2024 · similarity 0.87)
- SNIP: Bridging Mathematical Symbolic and Numeric Realms with Unified Pre-training (ICLR 2024 · similarity 0.86)
- Position: Algebra Unveils Deep Learning - An Invitation to Neuroalgebraic Geometry (ICML 2025 · similarity 0.86)
- Computational Algebra with Attention: Transformer Oracles for Border Basis Algorithms (NeurIPS 2025 · similarity 0.85)
- Mastering Symbolic Operations: Augmenting Language Models with Compiled Neural Networks (ICLR 2024 · similarity 0.84)
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.