Research map / Probabilistic and scientific modeling / Molecules and proteins
Multi-Objective Molecule Generation using Interpretable Substructures
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 |
|---|---|---|---|
| Use rationale-conditioned generation instead of REINVENT | Multi-objective molecule generation · GSK3β + JNK3 + QED + SA | Success: 47.9% -> 74.8% | Helped |
| Use rationale-conditioned generation instead of REINVENT | Multi-objective molecule generation · GSK3β + JNK3 | Novelty: 39.7% -> 97.3% | Helped |
| Use rationale-conditioned generation instead of REINVENT | Multi-objective molecule generation · GSK3β + JNK3 + QED + SA | Diversity: 0.621 -> 0.701 | Helped |
| Use rationale-conditioned generation instead of REINVENT | Molecule generation · GSK3β | Success: 99.3% -> 100% | No clear effect |
| Use rationale-conditioned generation instead of REINVENT | Molecule generation · GSK3β | Diversity: 0.733 -> 0.888 | Hurt |
| Use rationale-conditioned generation instead of REINVENT | Molecule generation · GSK3β + JNK3 | Diversity: 0.595 -> 0.824 | Hurt |
| Add molecular rationales to the graph generator rather than generating from scratch (ablation) | GVAE-RL graph VAE · GSK3β + JNK3 | Success: 40.7% -> 100% | Helped |
| Add molecular rationales to the graph generator rather than generating from scratch (ablation) | GVAE-RL graph VAE · GSK3β + JNK3 + QED + SA | Success: 2.1% -> 74.8% | Helped |
| Use MCTS rationale extraction instead of Integrated Gradient | Toxicity rationale identification · Toxicity dataset | Exact Match: 39.4% -> 46.0% | Helped |
| Use MCTS rationale extraction instead of Integrated Gradient | Toxicity rationale identification · Toxicity dataset | Partial Match AUC: 0.857 -> 0.861 | Helped |
| Generate using rationales rather than REINVENT | Molecule generation · GSK3β + JNK3 | FCD: 36.75 -> 25.52 | Helped |
Closest papers by meaning
- Projecting Molecules into Synthesizable Chemical Spaces (ICML 2024 · similarity 0.88)
- Representing Molecules as Random Walks Over Interpretable Grammars (ICML 2024 · similarity 0.87)
- Procedural Synthesis of Synthesizable Molecules (ICLR 2025 · similarity 0.87)
- Domain-Agnostic Molecular Generation with Chemical Feedback (ICLR 2024 · similarity 0.87)
- Molecule Design by Latent Prompt Transformer (NeurIPS 2024 · similarity 0.87)
- Controllable 3D Molecular Generation for Structure-Based Drug Design Through Bayesian Flow Networks and Gradient Integration (NeurIPS 2025 · similarity 0.87)
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.