Composable Score-based Graph Diffusion Model for Multi-Conditional Molecular Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912611232120832 |
|---|---|
| author | Qiao, Anjie Wang, Zhen Chen, Chuan Lian, DeFu Chen, Enhong |
| author_facet | Qiao, Anjie Wang, Zhen Chen, Chuan Lian, DeFu Chen, Enhong |
| contents | Controllable molecular graph generation is essential for material and drug discovery, where generated molecules must satisfy diverse property constraints. While recent advances in graph diffusion models have improved generation quality, their effectiveness in multi-conditional settings remains limited due to reliance on joint conditioning or continuous relaxations that compromise fidelity. To address these limitations, we propose Composable Score-based Graph Diffusion model (CSGD), the first model that extends score matching to discrete graphs via concrete scores, enabling flexible and principled manipulation of conditional guidance. Building on this foundation, we introduce two score-based techniques: Composable Guidance (CoG), which allows fine-grained control over arbitrary subsets of conditions during sampling, and Probability Calibration (PC), which adjusts estimated transition probabilities to mitigate train-test mismatches. Empirical results on four molecular datasets show that CSGD achieves state-of-the-art performance, with a 15.3% average improvement in controllability over prior methods, while maintaining high validity and distributional fidelity. Our findings highlight the practical advantages of score-based modeling for discrete graph generation and its capacity for flexible, multi-property molecular design. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_09451 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Composable Score-based Graph Diffusion Model for Multi-Conditional Molecular Generation Qiao, Anjie Wang, Zhen Chen, Chuan Lian, DeFu Chen, Enhong Machine Learning Controllable molecular graph generation is essential for material and drug discovery, where generated molecules must satisfy diverse property constraints. While recent advances in graph diffusion models have improved generation quality, their effectiveness in multi-conditional settings remains limited due to reliance on joint conditioning or continuous relaxations that compromise fidelity. To address these limitations, we propose Composable Score-based Graph Diffusion model (CSGD), the first model that extends score matching to discrete graphs via concrete scores, enabling flexible and principled manipulation of conditional guidance. Building on this foundation, we introduce two score-based techniques: Composable Guidance (CoG), which allows fine-grained control over arbitrary subsets of conditions during sampling, and Probability Calibration (PC), which adjusts estimated transition probabilities to mitigate train-test mismatches. Empirical results on four molecular datasets show that CSGD achieves state-of-the-art performance, with a 15.3% average improvement in controllability over prior methods, while maintaining high validity and distributional fidelity. Our findings highlight the practical advantages of score-based modeling for discrete graph generation and its capacity for flexible, multi-property molecular design. |
| title | Composable Score-based Graph Diffusion Model for Multi-Conditional Molecular Generation |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2509.09451 |