arXiv 2510.12719

Multitask finetuning and acceleration of chemical pretrained models for small molecule drug property prediction

By Matthew Adrian, Yunsie Chung, et al.

Published 2025-10-14

Mindmap

Browse the paper's core ideas, clusters, and relationships in a structured outline.

Chemical pretrained models, sometimes referred to as foundation models, are receiving considerable interest for drug discovery applications. The general chemical knowledge extracted from self-supervised training has the potential to improve predictions for critical drug discovery endpoints, including on-target potency and ADMET properties. Multi-task learning has previously been successfully leveraged to improve pre…

View the original paper on arXiv