# AromaGen: Interactive Generation of Rich Olfactory Experiences with Multimodal Language Models

DevFeed: [AromaGen: Interactive Generation of Rich Olfactory Experiences with Multimodal Language Models](<https://devfeed.tech/articles/aromagen-interactive-generation-of-rich-olfactory-experiences-with-multimodal-language-models-118207.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2604.01650>)

Author: Awu Chen, Silang Wang, Yunge Wen, Jianing Yu, Zinab Eisa, Ana Julia Santos Lopes, Jas Brooks, Hiroshi Ishii, Paul Pu Liang

Published: 2026-11-05T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [human computer interaction](<https://devfeed.tech/topics/human-computer-interaction.md>), [AI for interactive applications](<https://devfeed.tech/topics/ai-for-interactive-applications.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Retrieval Augmented Generation (RAG)](<https://devfeed.tech/topics/retrieval-augmented-generation-rag.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-powered](<https://devfeed.tech/tags/ai-powered.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [channel](<https://devfeed.tech/tags/channel.md>), [choice](<https://devfeed.tech/tags/choice.md>), [composition](<https://devfeed.tech/tags/composition.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-hc](<https://devfeed.tech/tags/cs-hc.md>), [data](<https://devfeed.tech/tags/data.md>), [experience](<https://devfeed.tech/tags/experience.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [food](<https://devfeed.tech/tags/food.md>), [generation](<https://devfeed.tech/tags/generation.md>), [interactive](<https://devfeed.tech/tags/interactive.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [language](<https://devfeed.tech/tags/language.md>), [language-models](<https://devfeed.tech/tags/language-models.md>), [maps](<https://devfeed.tech/tags/maps.md>), [memory](<https://devfeed.tech/tags/memory.md>), [models](<https://devfeed.tech/tags/models.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [multimodal-ai](<https://devfeed.tech/tags/multimodal-ai.md>)

## AI overview

AromaGen maps free-form natural-language descriptions to compositions of 12 base odorants in a wearable olfactory system, with natural-language feedback for iterative refinement. In a between-subjects study of 60 participants testing 50 real-world benchmark smells, participants identified generated target smells above chance in a three-alternative forced-choice task. Retrieval-augmented AI compositions performed comparably to both human-expert and zero-shot AI compositions. Feedback significantly improved perceived similarity for AI and human-expert compositions. These results support the feasibility of generating perceptible smell experiences from language using pretrained multimodal AI and human aroma-composition data.

## Source excerpt

arXiv:2604.01650v5 Announce Type: replace-cross Abstract: Smell's connections with food, memory, and social experience have motivated researchers to bring olfaction into interactive systems. Multimodal AI opens new possibilities for generating smells from natural language, yet it remains unclear whether pretrained models encode sufficient olfactory knowledge to translate language into perceptible compositions. We present AromaGen, an AI-powered 12-channel wearable olfactory system that maps free-form natural-language descriptions to 12 base odorants selected to cover a semantically derived olfactory space, while supporting iterative refinement through natural-language feedback. In a between-subjects study (N=60) across 50 real-world benchmark smells, participants distinguished AromaGen-generated target smells above chance in a three-alternative forced-choice task, with retrieval-augmented AI composition performing comparably to human-expert composition and zero-shot AI composition. Natural-language feedback significantly improved the perceived similarity of both AI- and human-expert compositions. Our findings demonstrate the feasibility of using pretrained multimodal AI, with human aroma composition data, to generate perceptible olfactory experiences from natural language.