{
  "abstract": "Most image-generation systems produce images through learned transformations of latent representations, iterative denoising, autoregressive prediction, or combinations of these processes. An alternative design perspective is to represent image generation as a sequence of deposition operations in which new visual content is progressively introduced into an evolving image. This paper proposes Adaptive Layered Deposition for Image Generation (ALD-IG), a conceptual and algorithmic framework that formalizes image synthesis as state-dependent construction through variable numbers of spatially localized layers. At each step, a controller evaluates the current image state and selects a deposition action, including adding new content, modifying an existing region, replacing previously generated content, adjusting inter-layer relationships, or terminating generation. Each operation is evaluated according to its expected contribution to image quality and its computational cost. Unlike fixed-depth layering, the proposed framework permits adaptive stopping and non-monotonic image updates. A mathematical formulation is developed using an evolving image state, layer representations, deposition policies, compositing operators, and a quality–cost objective. A practical implementation strategy, training procedure, benchmark design, and ablation study are also specified. The framework is intended to complement existing generative architectures rather than presuppose their replacement. Its central research hypothesis is that state-dependent deposition and selective revision may improve controllability or computational efficiency under appropriate conditions. Since layered image synthesis, progressive generation, and iterative refinement have already been investigated, the proposed contribution must be evaluated specifically against these related approaches. No experimental superiority is claimed in this paper; empirical validation remains future work.",
  "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/41mdrecwtjxb0.bib",
  "cited_by": 0,
  "doi": "10.5281/zenodo.23556189",
  "fulltext_url": null,
  "id": "41mdrecwtjxb0",
  "issns": [],
  "landing_url": "https://zenodo.org/records/23556189",
  "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/23556189/files/Adaptive%20Layered%20Deposition%20for%20Image%20Generation.pdf/content",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-22",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/41mdrecwtjxb0.ris",
  "sources": [
    {
      "first_seen": "2026-11-05T12:20:44Z",
      "native_id": "23556189",
      "source": "zenodo",
      "url": "https://zenodo.org/records/23556189"
    }
  ],
  "subjects": [],
  "title": "Adaptive Layered Deposition for Image Generation: A State-Dependent Framework for Progressive Construction, Selective Modification, and Variable-Depth Synthesis",
  "title_text": "Adaptive Layered Deposition for Image Generation: A State-Dependent Framework for Progressive Construction, Selective Modification, and Variable-Depth Synthesis",
  "type": "preprint",
  "updated_at": "2026-11-05T12:20:44Z",
  "url": "https://research.draconitic.com/paper/41mdrecwtjxb0",
  "venue": "Zenodo",
  "versions": []
}