{
  "abstract": "Sampling latency compounds in diffusion workflows, where users generate and discard many candidates before keeping one. Surprisingly, the poor outputs of standard few-step samplers do not reflect a lack of reconstruction capacity: by optimizing only the initial noise, frozen 3-4-step samplers can closely reproduce their corresponding full-step outputs. Building on this finding, we learn corrections to the initial noise and denoising updates using endpoint supervision, improving correspondence with full-step outputs generated from the same noise and prompt. The resulting previews allow users to screen candidates cheaply and reserve full-step generation for promising ones. Input correction also transfers across sampling budgets without retraining. Experiments show substantial improvements in reference fidelity, including 53-78% lower reconstruction MSE than retrained LD3 on unconditional benchmarks, alongside improved ranking preservation and candidate selection on SD1.5, SDXL, and FLUX.1-dev.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.34406",
  "author_total": 3,
  "authors": [
    {
      "name": "Jing Jia"
    },
    {
      "name": "Sifan Liu"
    },
    {
      "name": "Guanyang Wang"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/732sjerark17m.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.34406",
  "fulltext_url": null,
  "id": "732sjerark17m",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.34406",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.CV"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1707|Generative Adversarial Networks and Image Synthesis"
    ]
  ],
  "openalex": "W7214747697",
  "pdf_url": "https://arxiv.org/pdf/2609.34406",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-28",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/732sjerark17m.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:25:26Z",
      "native_id": "2609.34406",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.34406"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "Unlocking Few-Step Diffusion for Faithful Previews",
  "title_text": "Unlocking Few-Step Diffusion for Faithful Previews",
  "type": "preprint",
  "updated_at": "2026-10-30T12:25:26Z",
  "url": "https://research.draconitic.com/paper/732sjerark17m",
  "venue": "arXiv",
  "versions": []
}