Large Language Models for Controllable Multi-property Multi-objective Molecule Optimization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dey, Vishal, Hu, Xiao, Ning, Xia
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915313104191488
author Dey, Vishal
Hu, Xiao
Ning, Xia
author_facet Dey, Vishal
Hu, Xiao
Ning, Xia
contents In real-world drug design, molecule optimization requires selectively improving multiple molecular properties up to pharmaceutically relevant levels, while maintaining others that already meet such criteria. However, existing computational approaches and instruction-tuned LLMs fail to capture such nuanced property-specific objectives, limiting their practical applicability. To address this, we introduce C-MuMOInstruct, the first instruction-tuning dataset focused on multi-property optimization with explicit, property-specific objectives. Leveraging C-MuMOInstruct, we develop GeLLMO-Cs, a series of instruction-tuned LLMs that can perform targeted property-specific optimization. Our experiments across 5 in-distribution and 5 out-of-distribution tasks show that GeLLMO-Cs consistently outperform strong baselines, achieving up to 126% higher success rate. Notably, GeLLMO-Cs exhibit impressive 0-shot generalization to novel optimization tasks and unseen instructions. This offers a step toward a foundational LLM to support realistic, diverse optimizations with property-specific objectives. C-MuMOInstruct and code are accessible through https://github.com/ninglab/GeLLMO-C.
format Preprint
id arxiv_https___arxiv_org_abs_2505_23987
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Large Language Models for Controllable Multi-property Multi-objective Molecule Optimization
Dey, Vishal
Hu, Xiao
Ning, Xia
Machine Learning
Artificial Intelligence
Computation and Language
Biomolecules
In real-world drug design, molecule optimization requires selectively improving multiple molecular properties up to pharmaceutically relevant levels, while maintaining others that already meet such criteria. However, existing computational approaches and instruction-tuned LLMs fail to capture such nuanced property-specific objectives, limiting their practical applicability. To address this, we introduce C-MuMOInstruct, the first instruction-tuning dataset focused on multi-property optimization with explicit, property-specific objectives. Leveraging C-MuMOInstruct, we develop GeLLMO-Cs, a series of instruction-tuned LLMs that can perform targeted property-specific optimization. Our experiments across 5 in-distribution and 5 out-of-distribution tasks show that GeLLMO-Cs consistently outperform strong baselines, achieving up to 126% higher success rate. Notably, GeLLMO-Cs exhibit impressive 0-shot generalization to novel optimization tasks and unseen instructions. This offers a step toward a foundational LLM to support realistic, diverse optimizations with property-specific objectives. C-MuMOInstruct and code are accessible through https://github.com/ninglab/GeLLMO-C.
title Large Language Models for Controllable Multi-property Multi-objective Molecule Optimization
topic Machine Learning
Artificial Intelligence
Computation and Language
Biomolecules
url https://arxiv.org/abs/2505.23987