Research map / Graph learning / Graph self-supervised learning
Differentiable Scaffolding Tree for Molecule Optimization
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 |
|---|---|---|---|
| Replace random local edits with gradient-based DST optimization and sampling (ablation) | DST · JNK3 | APS: 0.413 -> 0.928 | Helped |
| Replace random local edits with gradient-based DST optimization and sampling (ablation) | DST · GSK3β | APS: 0.455 -> 0.869 | Helped |
| Replace random local edits with gradient-based DST optimization and sampling (ablation) | DST · LogP | APS: 36.1 -> 47.1 | Helped |
| Replace random local edits with gradient-based DST optimization and sampling (ablation) | DST · JNK3+GSK3β | APS: 0.622 -> 0.827 | Helped |
| Replace random local edits with gradient-based DST optimization and sampling (ablation) | DST · QED+SA+JNK3+GSK3β | APS: 0.575 -> 0.752 | Helped |
| Use DST gradient-based optimization rather than the strongest reported baseline | Molecular optimization · JNK3 | APS: 0.803 -> 0.928 | Helped |
| Use DST gradient-based optimization rather than the strongest reported baseline | Molecular optimization · GSK3β | APS: 0.810 -> 0.869 | Helped |
| Use DST gradient-based optimization rather than the strongest reported baseline | Molecular optimization · LogP | APS: 44.1 -> 47.1 | Helped |
| Use DST gradient-based optimization rather than the strongest reported baseline | Molecular optimization · JNK3+GSK3β | APS: 0.795 -> 0.827 | Helped |
| Use DST gradient-based optimization rather than the strongest reported baseline | Molecular optimization · QED+SA+JNK3+GSK3β | APS: 0.632 -> 0.752 | Helped |
Closest papers by meaning
- Learning to Extend Molecular Scaffolds with Structural Motifs (ICLR 2022 · similarity 0.89)
- InversionGNN: A Dual Path Network for Multi-Property Molecular Optimization (ICLR 2025 · similarity 0.86)
- Representing Molecules as Random Walks Over Interpretable Grammars (ICML 2024 · similarity 0.86)
- Molecular Optimization Model with Patentability Constraint (AAAI 2024 · similarity 0.86)
- Gradual Optimization Learning for Conformational Energy Minimization (ICLR 2024 · similarity 0.86)
- Foundation Molecular Grammar: Multi-Modal Foundation Models Induce Interpretable Molecular Graph Languages (ICML 2025 · similarity 0.85)
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.