arXiv cs.AI· Awu Chen, Silang Wang, Yunge Wen, Jianing Yu, Zinab Eisa, Ana Julia Santos Lopes, Jas Brooks, Hiroshi Ishii, Paul Pu Liang·· 4 小时前AI 评分39
AromaGen：用多模态语言模型交互式生成丰富嗅觉体验
AromaGen: Interactive Generation of Rich Olfactory Experiences with Multimodal Language Models
AI 导读
研究者发布 AromaGen，一套由多模态 AI 驱动的 12 通道可穿戴嗅觉系统，可将自由自然语言描述映射到覆盖语义嗅觉空间的 12 种基础气味剂，并支持通过自然语言反馈迭代调整。在 N=60 的组间实验中，50 种真实气味基准下，参与者能在三选一任务中超过随机水平辨识出目标气味；检索增强的 AI 配方表现与人类专家及零样本 AI 配方相当，自然语言反馈显著提升了配方的感知相似度。
正文 · 原文
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.
| Subjects: | Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2604.01650 [cs.HC] |
| (or arXiv:2604.01650v5 [cs.HC] for this version) | |
| https://doi.org/10.48550/arXiv.2604.01650 arXiv-issued DOI via DataCite |
Submission history
From: Awu Chen [view email]
[v1]
Thu, 2 Apr 2026 05:47:07 UTC (6,228 KB)
[v2]
Thu, 23 Apr 2026 23:14:51 UTC (6,228 KB)
[v3]
Sat, 26 Sep 2026 04:00:39 UTC (10,783 KB)
[v4]
Fri, 9 Oct 2026 21:06:06 UTC (10,784 KB)
[v5]
Tue, 3 Nov 2026 19:28:57 UTC (10,785 KB)
来源：arXiv cs.AI · arxiv.org