Learning Perceptual Geometry of Natural Odor Mixtures from LLM-Derived Descriptors

Shared on Bluesky by 2 AI experts

• T Tomer Ullman @tomerullman.bsky.social amplified

  @neurovenki.bsky.social

  Excited to share this work with Farhad Pashakhanloo, where we queried LLMs to derive the perceptual geometry of natural odor mixtures! We propose a method to predict perceptual properties from molecular composition & eve…

  View on Bluesky →
• A AI Firehose @ai-firehose.column.social: Harvard researchers have used large language models to analyze connections between molecular composition and the perception of natural odors… →