MolDeTox: Evaluating Language Model's Stepwise Fragment Editing for Molecular Detoxification

TL;DR

MolDeTox introduces a new benchmark for evaluating molecular detoxification using stepwise fragment editing, improving the assessment of toxicity-aware molecular optimization in drug discovery.

Contribution

The paper presents MolDeTox, a novel, detailed benchmark for toxicity-aware molecular optimization that emphasizes fragment-level understanding and evaluation.

Findings

Understanding molecules at the fragment level improves structural validity.

MolDeTox enables more reliable evaluation of toxicity reduction methods.

Analysis reveals strengths and limitations of current models in detoxification tasks.

Abstract

Large Language Models (LLMs) and Vision Language Models (VLMs) have recently shown promising capabilities in various scientific domain. In particular, these advances have opened new opportunities in drug discovery, where the ability to understand and modify molecular structures is critical for optimizing drug properties such as efficacy and toxicity. However, existing models and benchmarks often overlook toxicity-related challenges, focusing primarily on general property optimization without adequately addressing safety concerns. In addition, even existing toxicity repair benchmarks suffer from limited data diversity, low structural validity of generated molecules, and heavy reliance on proxy models for toxicity assessment. To address these limitations, we propose MolDeTox, a novel benchmark for molecular detoxification, designed to enable fine-grained and reliable evaluation of toxicity-aware molecular optimization across stepwise tasks. We evaluate a wide range of general-purpose LLMs and VLMs under diverse settings, and demonstrate that understanding and generating molecules at the fragment-level improves structural validity and enhances the quality of generated molecules. Moreover, through detailed task-level performance analysis, MolDeTox provides an interpretable benchmark that enables a deeper understanding of the detoxification process. Our dataset is available at : https://huggingface.co/datasets/MolDeTox/MolDeTox