cs.LGNov 16, 2026

Classical structural priors improve neural molecular representations in molecule language models

Authors: Donghyeok Lee, Sangkwon Park, Munyoung Chang, Sungroh Yoon

Organizations: Interdisciplinary Program in Artificial Intelligence, Seoul National University, Seoul 08826, Republic of Korea · Department of Electrical and Computer Engineering, Seoul National University, Seoul 08826, Republic of Korea · Education and Research Program for Future ICT Pioneers, Seoul National University, Seoul 08826, Republic of Korea · Department of Otorhinolaryngology-Head & Neck Surgery, Seoul National University Bundang Hospital, Seongnam 13620, Republic of Korea

Abstract

Classical molecular similarity is central to chemical information systems, yet molecule-language models trained on paired structure-text data do not necessarily organize molecular embeddings according to graded structural relationships. Here we introduce Structure-Aware Guided Alignment (SAGA), a training-only regularizer that uses Centered Kernel Alignment to align neural molecular representations with fingerprint-based Tanimoto similarity, without adding parameters or requiring fingerprints at inference. In MolCA and Atomas, SAGA improves molecule-text retrieval, shifts non-exact retrieval toward closer structural analogues, and increases average scaffold-split property-prediction performance, while molecular description and text-to-molecule generation are improved or preserved in single-run ablations. These results show that established chemoinformatics priors can make molecule-language representations more chemically faithful while retaining their multimodal utility.

Figures & tables

Explore similar work

CardsList
  1. Retrieval-guided Twin Fusion with Similarity-aware Contrast for Molecule-Text Alignment

    Aug 17, 2026Shunshun Gu, Shengqi Qiu, Hang Zhou +1Molecular Representation LearningMultimodal Contrastive Learning

  2. MolE-RAG: Molecular Structure-Enhanced Retrieval-Augmented Generation for Chemistry

    Jun 4, 2026Joey Chan, Wonbin Kweon, Ashley Shin +4Retrieval-Augmented GenerationLLM Inference

  3. MolReps: Taking SELFIES, Making SMILES, Teaching LLMs

    Nov 16, 2026Ken Deng, Garrett MorrisChemical Language ModelsMolecular Representation Learning