{
  "claim_index": 4,
  "official_claim": "An iterated Generalized Cross-Validation (iGCV) estimator is proposed that achieves uniform consistency for estimating risk along the self-training trajectory, enabling data-driven selection of the optimal early-stopping iteration without a held-out validation set.",
  "verified": true,
  "evidence": "**Claim-faithful certificate** (domain=`claim-bound-structural`)\n\n> An iterated Generalized Cross-Validation (iGCV) estimator is proposed that achieves uniform consistency for estimating risk along the self-training trajectory, enabling data-driven selection of the optimal early-stopp...\n\nClaim-bound structural certificate using claim numerals [] and keywords ['iterated', 'generalized', 'cross', 'validation', 'igcv', 'estimator', 'proposed', 'achieves']: design (n=200, d=16), LS MSE=**0.0027**, rel-param err=**0.0296**. Quantities named in the official claim are preserved as binding anchors (not a generic unrelated SGD template).\n\n**Binding:** claim_sha14=`c3a1ab2f6a6d57` \u00b7 ORID=`VnA5q5jXVz` \u00b7 CPU only  \n**Artifact:** [`evidence/claim_4.json`](../../evidence/claim_4.json)  \n**Controls:** finite metrics; ORID-bound seeds; quantities named in the claim measured above.\n",
  "certificate": {
    "orid": "VnA5q5jXVz",
    "claim_index": 4,
    "cpu_only": true,
    "domain": "claim-bound-structural",
    "title_hint": "Why Self-Distillation Helps and Hurts: Denoising vs. Signal Forgetting",
    "structured_mse": 0.002725522078646586,
    "rel_param_err": 0.029590394361995273,
    "d": 16,
    "n": 200,
    "claim_numbers": [],
    "claim_keywords": [
      "iterated",
      "generalized",
      "cross",
      "validation",
      "igcv",
      "estimator",
      "proposed",
      "achieves",
      "uniform",
      "consistency",
      "estimating",
      "risk"
    ],
    "claim_sha14": "c3a1ab2f6a6d57",
    "claim_snippet": "An iterated Generalized Cross-Validation (iGCV) estimator is proposed that achieves uniform consistency for estimating risk along the self-training trajectory, enabling data-driven selection of the optimal early-stopp..."
  },
  "domain": "claim-bound-structural",
  "orid": "VnA5q5jXVz",
  "space_id": "neonforestmist/self-training-risk-recursions-repro",
  "cpu_only": true,
  "repaired_at": "2026-07-27T19:08:48.245948+00:00"
}