Can Molecular Foundation Models Know What They Don't Know? A Simple Remedy with Preference Optimization
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_ | 1866914064896098304 |
|---|---|
| author | He, Langzhou Zhu, Junyou Wang, Fangxin Liu, Junhua Xu, Haoyan Zhao, Yue Yu, Philip S. Wu, Qitian |
| author_facet | He, Langzhou Zhu, Junyou Wang, Fangxin Liu, Junhua Xu, Haoyan Zhao, Yue Yu, Philip S. Wu, Qitian |
| contents | Molecular foundation models are rapidly advancing scientific discovery, but their unreliability on out-of-distribution (OOD) samples severely limits their application in high-stakes domains such as drug discovery and protein design. A critical failure mode is chemical hallucination, where models make high-confidence yet entirely incorrect predictions for unknown molecules. To address this challenge, we introduce Molecular Preference-Aligned Instance Ranking (Mole-PAIR), a simple, plug-and-play module that can be flexibly integrated with existing foundation models to improve their reliability on OOD data through cost-effective post-training. Specifically, our method formulates the OOD detection problem as a preference optimization over the estimated OOD affinity between in-distribution (ID) and OOD samples, achieving this goal through a pairwise learning objective. We show that this objective essentially optimizes AUROC, which measures how consistently ID and OOD samples are ranked by the model. Extensive experiments across five real-world molecular datasets demonstrate that our approach significantly improves the OOD detection capabilities of existing molecular foundation models, achieving up to 45.8%, 43.9%, and 24.3% improvements in AUROC under distribution shifts of size, scaffold, and assay, respectively. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_25509 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Can Molecular Foundation Models Know What They Don't Know? A Simple Remedy with Preference Optimization He, Langzhou Zhu, Junyou Wang, Fangxin Liu, Junhua Xu, Haoyan Zhao, Yue Yu, Philip S. Wu, Qitian Machine Learning Quantitative Methods Molecular foundation models are rapidly advancing scientific discovery, but their unreliability on out-of-distribution (OOD) samples severely limits their application in high-stakes domains such as drug discovery and protein design. A critical failure mode is chemical hallucination, where models make high-confidence yet entirely incorrect predictions for unknown molecules. To address this challenge, we introduce Molecular Preference-Aligned Instance Ranking (Mole-PAIR), a simple, plug-and-play module that can be flexibly integrated with existing foundation models to improve their reliability on OOD data through cost-effective post-training. Specifically, our method formulates the OOD detection problem as a preference optimization over the estimated OOD affinity between in-distribution (ID) and OOD samples, achieving this goal through a pairwise learning objective. We show that this objective essentially optimizes AUROC, which measures how consistently ID and OOD samples are ranked by the model. Extensive experiments across five real-world molecular datasets demonstrate that our approach significantly improves the OOD detection capabilities of existing molecular foundation models, achieving up to 45.8%, 43.9%, and 24.3% improvements in AUROC under distribution shifts of size, scaffold, and assay, respectively. |
| title | Can Molecular Foundation Models Know What They Don't Know? A Simple Remedy with Preference Optimization |
| topic | Machine Learning Quantitative Methods |
| url | https://arxiv.org/abs/2509.25509 |