{
  "abstract": "Multimodal large language models (MLLMs) increasingly integrate text, images, and audio, yet taste remains an underrepresented modality in smartphone-centered artificial intelligence. Existing research has developed electronic tongues, portable taste interfaces, digital taste stimulation, and sensor–actuator systems for remote tasting. However, the integration of these capabilities into a unified smartphone-centered multimodal large language model presents additional research questions concerning sensory representation, semantic reasoning, taste generation, device control, and feedback-based adaptation. This paper proposes Smartphone-Native Bidirectional Gustatory Intelligence (SNBGI), a conceptual framework that connects gustatory input and gustatory output through a common multimodal reasoning architecture. The proposed system receives taste-related chemical measurements from electronic-tongue sensors or compatible sensing peripherals, interprets these observations in context, generates structured gustatory output specifications, and communicates with a device capable of producing controlled taste stimuli. A closed-loop mechanism enables the system to compare intended taste profiles with subsequent sensor measurements and user feedback. The framework distinguishes objective chemical sensing from subjective human taste perception and distinguishes smartphone software from the external hardware required for physical taste stimulation. We formulate research questions, mathematical objectives, implementation components, and an experimental protocol for evaluating taste classification, generation accuracy, perceptual correspondence, response latency, portability, and safety. The contribution is a proposed architecture and validation program, not a claim that digital taste sensing or generation has been invented for the first time. The framework aims to provide a foundation for investigating how smartphone-based multimodal AI can perceive, reason about, and generate taste-related experiences.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": null,
  "author_total": 1,
  "authors": [
    {
      "name": "Jincheng Zhang",
      "orcid": "0009-0005-1860-0009"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/17zrk5b4ph7km.bib",
  "cited_by": 0,
  "doi": "10.5281/zenodo.23494189",
  "fulltext_url": null,
  "id": "17zrk5b4ph7km",
  "issns": [],
  "landing_url": "https://zenodo.org/records/23494189",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [],
  "openalex": null,
  "pdf_url": "https://zenodo.org/api/records/23494189/files/Smartphone-Native%20Bidirectional%20Gustatory%20Multimodal%20Large%20Language%20Models.pdf/content",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-20",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/17zrk5b4ph7km.ris",
  "sources": [
    {
      "first_seen": "2026-11-03T12:21:18Z",
      "native_id": "23494189",
      "source": "zenodo",
      "url": "https://zenodo.org/records/23494189"
    }
  ],
  "subjects": [],
  "title": "Smartphone-Native Bidirectional Gustatory Multimodal Large Language Models: A Unified Framework for Taste Sensing, Reasoning, Generation, and Physical Taste Feedback",
  "title_text": "Smartphone-Native Bidirectional Gustatory Multimodal Large Language Models: A Unified Framework for Taste Sensing, Reasoning, Generation, and Physical Taste Feedback",
  "type": "preprint",
  "updated_at": "2026-11-03T12:21:18Z",
  "url": "https://research.draconitic.com/paper/17zrk5b4ph7km",
  "venue": "Zenodo",
  "versions": []
}