arXiv subjects

Laura Sisson

Publications and source records attributed to Laura Sisson.

4 recordsLinked to original sources

arXiv2026

Cultural Evolution of Perfumes since 1900

Perfumes are cultural artifacts and works of sensory art, composed from a finite, recombinable palette of notes that together evoke a distinctive scent impression. Here, we assemble the largest perfume corpus compiled to date, spanning multiple independent databases from 1900 to 2024, and study its evolution through a multidisciplinary computational framework. We first characterize perfumes by properties such as complexity and novelty using crowd-sourced data, finding that compositions have grown more minimalist in their scent profiles, yet more novel in their note combinations, a transition that began in the 1990s. Next, we construct copy-lineage networks and examine how notes are selected across successive time windows. We show that although imitation is pervasive, a growing share of notes drifts free of selection, and original creations retain a measurable quality premium, where they last longer, project further, and earn higher regard. Finally, we construct the collaboration network of master perfumers and show that a perfumer's creative style behaves as a social contagion, transmitted through collaboration and decaying with social distance. We discuss the principal forces shaping this evolution, including regulatory restrictions, cultural change, and the consolidation in the industry. Our findings position perfume as a culturally evolving system, akin to music and fashion, through which societies express and communicate hedonic sensory experiences.

physics.soc-ph↗

arXiv2026

Benchmark for Assessing Olfactory Perception of Large Language Models

Here we introduce the Olfactory Perception (OP) benchmark, designed to assess the capability of large language models (LLMs) to reason about smell. The benchmark contains 1,010 questions across eight task categories spanning odor classification, odor primary descriptor identification, intensity and pleasantness judgments, multi-descriptor prediction, mixture similarity, olfactory receptor activation, and smell identification from real-world odor sources. Each question is presented in two prompt formats, compound names and isomeric SMILES, to evaluate the effect of molecular representations. Evaluating 21 model configurations across major model families, we find that compound-name prompts consistently outperform isomeric SMILES, with gains ranging from +2.4 to +18.9 percentage points (mean approx +7 points), suggesting current LLMs access olfactory knowledge primarily through lexical associations rather than structural molecular reasoning. The best-performing model reaches 64.4\% overall accuracy, which highlights both emerging capabilities and substantial remaining gaps in olfactory reasoning. We further evaluate a subset of the OP across 21 languages and find that aggregating predictions across languages improves olfactory prediction, with AUROC = 0.86 for the best performing language ensemble model. LLMs should be able to handle olfactory and not just visual or aural information.

cs.CL↗

arXiv2023

Olfactory Label Prediction on Aroma-Chemical Pairs

The application of deep learning techniques on aroma-chemicals has resulted in models more accurate than human experts at predicting olfactory qualities. However, public research in this domain has been limited to predicting the qualities of single molecules, whereas in industry applications, perfumers and food scientists are often concerned with blends of many molecules. In this paper, we apply both existing and novel approaches to a dataset we gathered consisting of labeled pairs of molecules. We present graph neural network models capable of accurately predicting the odor qualities arising from blends of aroma-chemicals, with an analysis of how variations in architecture can lead to significant differences in predictive power.

cs.LG↗

arXiv2022

Odor Descriptor Understanding through Prompting

Embeddings from contemporary natural language processing (NLP) models are commonly used as numerical representations for words or sentences. However, odor descriptor words, like "leather" or "fruity", vary significantly between their commonplace usage and their olfactory usage, as a result traditional methods for generating these embeddings do not suffice. In this paper, we present two methods to generate embeddings for odor words that are more closely aligned with their olfactory meanings when compared to off-the-shelf embeddings. These generated embeddings outperform the previous state-of-the-art and contemporary fine-tuning/prompting methods on a pre-existing zero-shot odor-specific NLP benchmark.

cs.LG↗