# CORE: Automatic Molecule Optimization Using Copy & Refine Strategy

arXiv: https://arxiv.org/abs/1912.05910 · HTML: https://mlsanity.com/paper/1912.05910/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Nov 2019.

## Built on by (most cited first, 27 of 27)

- [ADBench: Anomaly Detection Benchmark](https://mlsanity.com/paper/2206.09426/) (2022)
- [Machine Learning for Synthetic Data Generation: A Review](https://mlsanity.com/paper/2302.04062/) (2023)
- [A Survey on Deep Graph Generation: Methods and Applications](https://mlsanity.com/paper/2203.06714/) (2022)
- [MolGenSurvey: A Systematic Survey in Machine Learning Models for Molecule Design](https://mlsanity.com/paper/2203.14500/) (2022)
- [Differentiable Scaffolding Tree for Molecular Optimization](https://mlsanity.com/paper/2109.10469/) (2021)
- [HINT: Hierarchical Interaction Network for Trial Outcome Prediction Leveraging Web Data](https://mlsanity.com/paper/2102.04252/) (2021)
- [Artificial Intelligence in Drug Discovery: Applications and Techniques](https://mlsanity.com/paper/2106.05386/) (2021)
- [Controllable Data Generation by Deep Learning: A Review](https://mlsanity.com/paper/2207.09542/) (2022)
- [Network Representation Learning: From Preprocessing, Feature Extraction to Node Embedding](https://mlsanity.com/paper/2110.07582/) (2021)
- [Uncertainty Quantification on Clinical Trial Outcome Prediction](https://mlsanity.com/paper/2401.03482/) (2024)
- [ChemMLLM: Chemical Multimodal Large Language Model](https://mlsanity.com/paper/2505.16326/) (2025)
- [Interpretable Molecular Graph Generation via Monotonic Constraints](https://mlsanity.com/paper/2203.00412/) (2022)
- [Molecule Generation for Drug Design: a Graph Learning Perspective](https://mlsanity.com/paper/2202.09212/) (2022)
- [Disentangled Spatiotemporal Graph Generative Models](https://mlsanity.com/paper/2203.00411/) (2022)
- [Molecule Design by Latent Space Energy-Based Modeling and Gradual Distribution Shifting](https://mlsanity.com/paper/2306.14902/) (2023)
- [DrugCLIP: Contrastive Drug-Disease Interaction For Drug Repurposing](https://mlsanity.com/paper/2407.02265/) (2024)
- [TrialBench: Multi-Modal Artificial Intelligence-Ready Clinical Trial Datasets](https://mlsanity.com/paper/2407.00631/) (2024)
- [Gradient GA: Gradient Genetic Algorithm for Drug Molecular Design](https://mlsanity.com/paper/2502.09860/) (2025)
- [Machine Learning Applications for Therapeutic Tasks with Genomics Data](https://mlsanity.com/paper/2105.01171/) (2021)
- [InversionGNN: A Dual Path Network for Multi-Property Molecular Optimization](https://mlsanity.com/paper/2503.01488/) (2025)
- [An In-depth Summary of Recent Artificial Intelligence Applications in Drug Design](https://mlsanity.com/paper/2110.05478/) (2021)
- [Molecule optimization via multi-objective evolutionary in implicit chemical space](https://mlsanity.com/paper/2212.08826/) (2022)
- [Language Interaction Network for Clinical Trial Approval Estimation](https://mlsanity.com/paper/2405.06662/) (2024)
- [Molecular Attributes Transfer from Non-Parallel Data](https://mlsanity.com/paper/2111.15146/) (2021)
- [An open unified deep graph learning framework for discovering drug leads](https://mlsanity.com/paper/2301.03424/) (2022)
- [ModuLM: Enabling Modular and Multimodal Molecular Relational Learning with Large Language Models](https://mlsanity.com/paper/2506.00880/) (2025)
- [MolWorld: Molecule World Models for Actionable Molecular Optimization](https://mlsanity.com/paper/2605.08954/) (2026)

## Abstract (arXiv)

Molecule optimization is about generating molecule Y with more desirable properties based on an input molecule X. The state-of-the-art approaches partition the molecules into a large set of substructures S and grow the new molecule structure by iteratively predicting which substructure from S to add. However, since the set of available substructures S is large, such an iterative prediction task is often inaccurate especially for substructures that are infrequent in the training data. To address this challenge, we propose a new generating strategy called "Copy & Refine" (CORE), where at each step the generator first decides whether to copy an existing substructure from input X or to generate a new substructure, then the most promising substructure will be added to the new molecule. Combining together with scaffolding tree generation and adversarial training, CORE can significantly improve several latest molecule optimization methods in various measures including drug likeness (QED), dopamine receptor (DRD2) and penalized LogP. We tested CORE and baselines using the ZINC database and CORE obtained up to 11% and 21% relatively improvement over the baselines on success rate on the complete test set and the subset with infrequent substructures, respectively.