Odor Maps from the LLM-derived similarity scores
Yuki Harada, Manuel Aleixandre, Manabu Okumura, Takamichi Nakamoto
TLDR
This paper explores using LLMs to create odor maps by comparing LLM-derived odor similarities with human evaluations.
Key contributions
- Calculated pairwise distances of odor descriptors using LLMs and three distance measures.
- Statistically compared LLM-derived odor similarities with original human evaluation data.
- Extended the approach to infer similarity for odor names (ingredients).
- Generated an odor map of essential oils, showing proximity corresponds to human perception.
Why it matters
This research demonstrates the promising capability of large language models to understand and map complex odor relationships. It opens new avenues for computational olfaction, potentially aiding in fragrance design, food science, and understanding human perception of smell.
Original Abstract
The application of large language models (LLMs) to OdorSpace analysis attracts growing interest. Recent studies have explored the comparison of sensory evaluation spaces derived from LLMs with odor character profiles in the Dravnieks' dataset. In this study, we calculated pairwise distances of odor descriptors using three distance measures and statistically compared these LLM-derived similarities with distances derived from the original data. Next, we extended this approach to odor names (ingredients). Statistical comparison revealed that LLMs can infer odor similarity to some degree, suggesting the potential of odor maps generated from these similarity data. Applying this approach, we generated an odor map of essential oils. It demonstrates that essential oils within the same group are closely located in the odor map, suggesting that the proximity in the odor map corresponds to human evaluation.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.