{
  "abstract": "Masked prediction learns to infer missing variables from visible context selected by a mask schedule. When does small excess risk guarantee recovery of the true joint distribution? We study this question on finite product spaces under masked-block log loss, with conditionals induced by a single joint distribution. To quantify recovery, we introduce an $\\varepsilon$-identifiability modulus measuring the worst-case error among joint distributions with population excess risk at most $\\varepsilon$. For data with separated modes pinned down by sufficiently large visible contexts, schedules that always retain such contexts can permit nonvanishing mode-weight errors while incurring exponentially small excess risk. An information decomposition explains why: the loss captures mode-weight mismatch only where the visible context leaves the mode uncertain. We prove two-sided bounds showing that, over a fixed range of mode weights, sensitivity to mode reweighting is governed by residual mode uncertainty averaged over the mask schedule. Assigning schedule mass to low-visibility masks that retain this uncertainty yields recovery bounds within the reweighting family. Beyond this family, positive full-mask probability characterizes uniform control of joint KL divergence by excess risk. Exact calculations and controlled gradient optimization validate these predictions.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2608.01383",
  "author_total": 2,
  "authors": [
    {
      "name": "Yichao Cai"
    },
    {
      "name": "Javen Qinfeng Shi"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/9190w8arpz2k5.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2608.01383",
  "fulltext_url": null,
  "id": "9190w8arpz2k5",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2608.01383",
  "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.IT"
    ],
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "math.IT"
    ],
    [
      "arxiv",
      "stat.ML"
    ],
    [
      "openalex",
      "1702|Privacy-Preserving Technologies in Data"
    ]
  ],
  "openalex": "W7172395969",
  "pdf_url": "https://arxiv.org/pdf/2608.01383",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-05",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/9190w8arpz2k5.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:27:10Z",
      "native_id": "2608.01383",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2608.01383"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.ML",
      "subfield": "statistics"
    }
  ],
  "title": "An Identifiability Theory of Masked Prediction: Mode Blindness and Mask Schedules",
  "title_text": "An Identifiability Theory of Masked Prediction: Mode Blindness and Mask Schedules",
  "type": "preprint",
  "updated_at": "2026-10-30T12:27:16Z",
  "url": "https://research.draconitic.com/paper/9190w8arpz2k5",
  "venue": "arXiv",
  "versions": []
}