NOSE: Neural Olfactory-Semantic Embedding with Tri-Modal Orthogonal Contrastive Learning

TL;DR

NOSE is a novel multi-modal embedding framework that aligns molecular, receptor, and linguistic olfactory data, achieving state-of-the-art results and strong zero-shot generalization.

Contribution

It introduces a tri-modal orthogonal contrastive learning approach with a weak positive sampling strategy for comprehensive olfactory representation.

Findings

NOSE outperforms existing methods on olfactory tasks.

It demonstrates strong zero-shot generalization capabilities.

The framework effectively captures the biological and semantic aspects of olfaction.

Abstract

Olfaction lies at the intersection of chemical structure, neural encoding, and linguistic perception, yet existing representation methods fail to fully capture this pathway. Current approaches typically model only isolated segments of the olfactory pathway, overlooking the complete chain from molecule to receptors to linguistic descriptions. Such fragmentation yields learned embeddings that lack both biological grounding and semantic interpretability. We propose NOSE (Neural Olfactory-Semantic Embedding), a representation learning framework that aligns three modalities along the olfactory pathway: molecular structure, receptor sequence, and natural language description. Rather than simply fusing these signals, we decouple their contributions via orthogonal constraints, preserving the unique encoded information of each modality. To address the sparsity of olfactory language, we introduce a weak positive sample strategy to calibrate semantic similarity, preventing erroneous repulsion of similar odors in the feature space. Extensive experiments demonstrate that NOSE achieves state-of-the-art (SOTA) performance and excellent zero-shot generalization, confirming the strong alignment between its representation space and human olfactory intuition.Code and data are available at https://github.com/Xianyusyy/NOSE