{
  "abstract": "Existing three-dimensional generative approaches commonly construct or infer geometry through learned distributions, implicit neural representations, explicit primitives, or point-based structures. This paper proposes Phase-Transition-Driven 3D Generative Computing (PT3D-GC), a conceptual framework in which three-dimensional geometry emerges through controlled transitions between material-like computational states. Rather than treating a target object exclusively as a static collection of geometric primitives, PT3D-GC models generation as a spatially distributed process involving nucleation, phase-interface propagation, crystallization, solidification, phase separation, and structural locking.\n\nThe framework represents a three-dimensional domain using coupled fields describing phase state, material concentration, temperature-like control, crystallinity, and structural stability. A phase-transition controller determines when and where transformations occur, while a state-evolution engine updates the computational material over discrete time steps. The final geometry is extracted from the stabilized phase configuration using an isosurface or occupancy threshold. A hybrid implementation can combine phase-field dynamics with differentiable rendering, implicit neural representations, or explicit surface representations.\n\nThis paper formalizes the proposed framework, defines its governing equations and generation algorithm, and establishes an experimental protocol for evaluating geometric fidelity, structural diversity, computational cost, topological complexity, and process stability. Four principal modes—solidification, evaporation-driven crystallization, freezing, and phase separation—are described. Because the framework has not yet been experimentally validated, its potential advantages are presented as hypotheses rather than demonstrated results. The study aims to establish a testable direction for process-based three-dimensional generation in which state transitions serve as the primary mechanism of geometric construction.",
  "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/er2q1btzt1aa3.bib",
  "cited_by": 0,
  "doi": "10.5281/zenodo.23636460",
  "fulltext_url": null,
  "id": "er2q1btzt1aa3",
  "issns": [],
  "landing_url": "https://zenodo.org/records/23636460",
  "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/23636460/files/Phase-Transition-Driven%203D%20Generative%20Computing.pdf/content",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-25",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/er2q1btzt1aa3.ris",
  "sources": [
    {
      "first_seen": "2026-11-08T12:21:22Z",
      "native_id": "23636460",
      "source": "zenodo",
      "url": "https://zenodo.org/records/23636460"
    }
  ],
  "subjects": [],
  "title": "Phase-Transition-Driven 3D Generative Computing: A State-Evolution Framework for Constructing Three-Dimensional Geometry through Nucleation, Crystallization, Solidification, and Phase Separation",
  "title_text": "Phase-Transition-Driven 3D Generative Computing: A State-Evolution Framework for Constructing Three-Dimensional Geometry through Nucleation, Crystallization, Solidification, and Phase Separation",
  "type": "preprint",
  "updated_at": "2026-11-08T12:21:22Z",
  "url": "https://research.draconitic.com/paper/er2q1btzt1aa3",
  "venue": "Zenodo",
  "versions": []
}