arXiv：cs.LG· Vansh Ramani, Har Ashish Arora, Dhairya Kuchhal, Sayan Ranu, Tarak Karmakar·· 3 小时前AI 评分34
DISSOLVR：可解释且快速的水相与有机溶解度预测框架
DISSOLVR: An Interpretable and Fast Framework for Aqueous and Organic Solubility Prediction
AI 导读
DISSOLVR 是一个透明可解释的分子溶解度预测框架，通过将分子映射到物理基础描述符实现结构不变性，性能接近实验不确定性的随机极限并具备 OOD 泛化能力。该框架还引入 LLM 辅助的事后解释流程，将符号模型产物转化为化学叙事，并通过对 22 位化学专家的调研基准验证了其可解释性。论文已被 ICML 2026 接收。
正文
Abstract:High-fidelity solubility prediction is fundamental to pharmaceutical development and environmental partitioning, where accurate modeling must couple molecular structure with thermodynamic behavior across diverse chemical environments. However, recent advancements have been dominated by deep learning architectures that often sacrifice physical interpretability for predictive power. We challenge this trend by showing that state-of-the-art performance does not require such non-transparent architectures. To address this, we introduce DISSOLVR, a transparent framework for molecular solubility prediction. In addition, we perform a comprehensive literature review and a benchmarking study against various methods. We show that DISSOLVR approaches the aleatoric limit of experimental uncertainty and achieves OOD generalization through structural invariance, derived by mapping molecules to physically-grounded descriptors. Then, we present an LLM-assisted post-hoc explanation pipeline that bridges the gap between symbolic model artifacts and chemically grounded narratives. Finally, a comparative benchmark of a survey involving 22 expert chemists reveals that expert evaluators provide deep insights.
| Comments: | 48 pages, 19 tables, 6 figures. Accepted to the 43rd International Conference on Machine Learning (ICML 2026) |
| Subjects: | Machine Learning (cs.LG); Chemical Physics (physics.chem-ph) |
| Cite as: | arXiv:2610.02574 [cs.LG] |
| (or arXiv:2610.02574v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.02574 arXiv-issued DOI via DataCite (pending registration) |
|
| Journal reference: | Proceedings of the 43rd International Conference on Machine Learning, Seoul, South Korea. PMLR 306, 2026 |
Submission history
From: Har Ashish Arora [view email]
[v1]
Thu, 1 Oct 2026 23:10:05 UTC (7,762 KB)
来源：arXiv：cs.LG · arxiv.org