# Robust Molecular Property Prediction via Densifying Scarce Labeled Data

arXiv: https://arxiv.org/abs/2506.11877 · HTML: https://mlsanity.com/paper/2506.11877/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 16 references that are in the map.
- First posted on arXiv 13 Jun 2025.

## Builds on (most cited first, 8 of 8)

- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift](https://mlsanity.com/paper/1906.02530/) (2019)
- [Optimizing Millions of Hyperparameters by Implicit Differentiation](https://mlsanity.com/paper/1911.02590/) (2019)
- [Exploring Chemical Space with Score-based Out-of-distribution Generation](https://mlsanity.com/paper/2206.07632/) (2022)
- [Low Data Drug Discovery with One-shot Learning](https://mlsanity.com/paper/1611.03199/) (2016)
- [Drug Discovery under Covariate Shift with Domain-Informed Prior Distributions over Functions](https://mlsanity.com/paper/2307.15073/) (2023)
- [Set-based Meta-Interpolation for Few-Task Meta-Learning](https://mlsanity.com/paper/2205.09990/) (2022)

## Built on by (most cited first, 2 of 2)

- [Conformal Prediction for Molecular Properties under Label Shift](https://mlsanity.com/paper/2608.17678/) (2026)
- [Improving the accuracy and generalizability of molecular property regression models with a substructure-substitution-rule-informed framework](https://mlsanity.com/paper/2511.08314/) (2025)

## Abstract (arXiv)

A widely recognized limitation of molecular prediction models is their reliance on structures observed in the training data, resulting in poor generalization to out-of-distribution compounds. Yet in drug discovery, the compounds most critical for advancing research often lie beyond the training set, making the bias toward the training data particularly problematic. This mismatch introduces substantial covariate shift, under which standard deep learning models produce unstable and inaccurate predictions. Furthermore, the scarcity of labeled data-stemming from the onerous and costly nature of experimental validation-further exacerbates the difficulty of achieving reliable generalization. To address these limitations, we propose a novel bilevel optimization approach that leverages unlabeled data to interpolate between in-distribution (ID) and out-of-distribution (OOD) data, enabling the model to learn how to generalize beyond the training distribution. We demonstrate significant performance gains on challenging real-world datasets with substantial covariate shift, supported by t-SNE visualizations highlighting our interpolation method.