{
 "spxi_instrument": {
  "method": "0.9",
  "name": "Capture Registry (longitudinal observation)",
  "level": 0,
  "level_name": "Measurement",
  "url": "https://www.spxi.dev/methods/#m-0-9",
  "inventory": "https://www.spxi.dev/methods/",
  "statement": "SPXI instrument: method 0.9 of 87, level 0 (Measurement), Capture Registry (longitudinal observation).",
  "also": []
 },
 "slug": "ithaca-deepmind-aio-20261031",
 "date": "2026-10-31",
 "surface": "Google AI Overview",
 "surface_basis": "Operator default of 2026-10-01 (runs begin in the Overview unless he says AI Mode); the 'AI Mode Conversation' paste is the Overview expanded (operator, 2026-10-18).",
 "surfaces": [
  "Google AI Overview"
 ],
 "auth": "signed out, incognito",
 "auth_basis": "Operator default of 2026-10-14.",
 "ev": "paste",
 "mt": "ITHACA ON THE OVERVIEW — 62%, 25% TO 72%, 71%, WITHIN 30 YEARS, ALL HELD, WITH ITS ARCHITECTURE AND ITS LICENCE; THE MODEL'S OWN TERM (TO 'ASSIST AND EXPAND THE HISTORIAN'S WORKFLOW') CARRIED BY ITS FIGURES ALONE",
 "match": "BROAD MATCH",
 "match_note": null,
 "s": "Machine Reception",
 "q": "Ithaca DeepMind",
 "q_kind": "the common name of an entity of the Revelation First meta-aggregator, 'the field' (Ithaca; record https://www.revelationfirst.com/aggregators/ithaca/, compiled 2026-10-30), supplied for capture by the archive and run by the operator, to supply the composition layer's account of the entity for its record. The operator added the disambiguator to a name that is bare a homonym. A baseline: the record was built on revelationfirst.com the day before and no card cites it. NEW address.",
 "cites": 4,
 "cite_list": [
  {
   "n": 1,
   "site": "Google DeepMind",
   "rel": "third_party",
   "url": null,
   "title": "Predicting the past with Ithaca — Google DeepMind",
   "snip": "Core: Ithaca deep neural network (restores ancient Greek inscriptions, identifies geographic origins, establishes creation dates). Architecture: Sparse multi-he...",
   "note": null
  },
  {
   "n": 2,
   "site": "GitHub",
   "rel": "third_party",
   "url": null,
   "title": "GitHub - google-deepmind/ithaca: Restoring and attributing ancient texts using deep neural networks · GitHub",
   "snip": "Overview: Ithaca is a deep neural network developed by DeepMind and academic partners for restoring, geographically attributing, and chronologically dating anci...",
   "note": null
  },
  {
   "n": 3,
   "site": "Predicting the Past",
   "rel": "third_party",
   "url": null,
   "title": "Predicting the Past",
   "snip": "Contextualising, restoring, and attributing ancient texts Ithaca and Aeneas are generative AI models designed to assist historians in the study of ancient Greek...",
   "note": null
  },
  {
   "n": 4,
   "site": "Nature",
   "rel": "third_party",
   "url": null,
   "title": "Restoring and attributing ancient texts using deep neural networks | Nature",
   "snip": "Model: Ithaca, a deep neural network trained on ancient Greek inscriptions (7th century BC to 5th century AD) from the I.PHI dataset (78,608 texts), simultaneou...",
   "note": null
  }
 ],
 "archive_controlled_cites": 0,
 "transcript": "[Google AI Overview at the query, signed out and incognito (operator defaults of 2026-10-01 and 2026-10-14), received 2026-10-31 00:04 EDT. The paste is the Overview expanded (the 'AI Mode Conversation' view, with the source cards that follow the answer); the 'You said' line carries the query twice, run together, and is given once here. Inline links are Google redirect URLs, kept as pasted.]\n\n[QUERENT] Ithaca DeepMind\n\n[SYSTEM] Ithaca is a deep neural network developed by Google DeepMind and academic partners to restore, geographically attribute, and chronologically date damaged ancient Greek inscriptions. [[1](https://www.google.com/goto?url=CAESYQHrOzAV9bhPXwiPFAkk0WrtjwF69osLlHBUWNzzI69S1WtwZTqUCsUsFHJRmbHqvOTqawUMo9x8Q-JEjDS2uEGe-nwFB8K7bhK_9tMlb9RUSeY2GDejKtZrzQPcaK1YQ8A)]\nKey Capabilities\n   * Text Restoration: Restores missing or damaged characters with 62% accuracy on its own (raising human historian accuracy from 25% to 72% when collaborating). [[1](https://www.google.com/goto?url=CAESYQHrOzAV9bhPXwiPFAkk0WrtjwF69osLlHBUWNzzI69S1WtwZTqUCsUsFHJRmbHqvOTqawUMo9x8Q-JEjDS2uEGe-nwFB8K7bhK_9tMlb9RUSeY2GDejKtZrzQPcaK1YQ8A)]\n   * Geographical Attribution: Identifies the original geographic location across 84 regions with 71% accuracy. [[1](https://www.google.com/goto?url=CAESdQHrOzAV_0MWkpVTpN7uvNvXlOBaZ40iii-iOTMqzp09h7b0HxKyB-_B3anpWxAEboASgFYZPv2f1wL6NWVEbFQlf4ki9DS6MGXMox5jtuJ6YxNutsIntryMo_rhg8_NQ3HN7Mpl7kEKmaM9FC0kvaP9Eb726w), [2](https://www.google.com/goto?url=CAESagHrOzAVvUud25tCo9gLWQgmo4vGXjLWeVFxE5AU0Uki3Gg-oMDVXRFMODGH1hrVZkipV_2rQt9md-GkhXBTCM2McqHlXUfunJiUdmKCar8LAG0D4xDoHA6uzaN6yPDhb_m-D2SdWYGdhSk)]\n   * Chronological Dating: Dates inscriptions within 30 years of their true historical creation date. [[1](https://www.google.com/goto?url=CAESdQHrOzAV_0MWkpVTpN7uvNvXlOBaZ40iii-iOTMqzp09h7b0HxKyB-_B3anpWxAEboASgFYZPv2f1wL6NWVEbFQlf4ki9DS6MGXMox5jtuJ6YxNutsIntryMo_rhg8_NQ3HN7Mpl7kEKmaM9FC0kvaP9Eb726w)]\nHow It Works\n   * Architecture: Uses joint character and word embeddings with stacked transformer decoder blocks and sparse multi-head attention. [[1](https://www.google.com/goto?url=CAESagHrOzAVvUud25tCo9gLWQgmo4vGXjLWeVFxE5AU0Uki3Gg-oMDVXRFMODGH1hrVZkipV_2rQt9md-GkhXBTCM2McqHlXUfunJiUdmKCar8LAG0D4xDoHA6uzaN6yPDhb_m-D2SdWYGdhSk), [2](https://www.google.com/goto?url=CAESdQHrOzAV_0MWkpVTpN7uvNvXlOBaZ40iii-iOTMqzp09h7b0HxKyB-_B3anpWxAEboASgFYZPv2f1wL6NWVEbFQlf4ki9DS6MGXMox5jtuJ6YxNutsIntryMo_rhg8_NQ3HN7Mpl7kEKmaM9FC0kvaP9Eb726w)]\n   * Training Data: Trained on the Packard Humanities Institute (PHI) Searchable Greek Inscriptions dataset. [[1](https://www.google.com/goto?url=CAESdQHrOzAV_0MWkpVTpN7uvNvXlOBaZ40iii-iOTMqzp09h7b0HxKyB-_B3anpWxAEboASgFYZPv2f1wL6NWVEbFQlf4ki9DS6MGXMox5jtuJ6YxNutsIntryMo_rhg8_NQ3HN7Mpl7kEKmaM9FC0kvaP9Eb726w), [2](https://www.google.com/goto?url=CAESYQHrOzAV9bhPXwiPFAkk0WrtjwF69osLlHBUWNzzI69S1WtwZTqUCsUsFHJRmbHqvOTqawUMo9x8Q-JEjDS2uEGe-nwFB8K7bhK_9tMlb9RUSeY2GDejKtZrzQPcaK1YQ8A)]\n   * Access & Open Source: The code is open-sourced under the Apache-2.0 license via the [GitHub - Google-DeepMind Ithaca Repository](https://www.google.com/goto?url=CAESYQHrOzAV9bhPXwiPFAkk0WrtjwF69osLlHBUWNzzI69S1WtwZTqUCsUsFHJRmbHqvOTqawUMo9x8Q-JEjDS2uEGe-nwFB8K7bhK_9tMlb9RUSeY2GDejKtZrzQPcaK1YQ8A), and interactive research tools can be explored through [Predicting the Past](https://www.google.com/goto?url=CAESVgHrOzAV5xjfiCSPWOfGgOXOl09A2Ia9t8w6n3GOcZKu8sNgrR6XHIbDPTs6hdVLhRmsBR_nK-WQVer7zY6gPSmm1pIQT5soqR4Dqyo3prIvXmk6EpMN). [[1](https://www.google.com/goto?url=CAESVgHrOzAV5xjfiCSPWOfGgOXOl09A2Ia9t8w6n3GOcZKu8sNgrR6XHIbDPTs6hdVLhRmsBR_nK-WQVer7zY6gPSmm1pIQT5soqR4Dqyo3prIvXmk6EpMN), [2](https://www.google.com/goto?url=CAESYQHrOzAV9bhPXwiPFAkk0WrtjwF69osLlHBUWNzzI69S1WtwZTqUCsUsFHJRmbHqvOTqawUMo9x8Q-JEjDS2uEGe-nwFB8K7bhK_9tMlb9RUSeY2GDejKtZrzQPcaK1YQ8A)]\nWould you like to know more about how Ithaca connects with newer models like Aeneas or how historians use the Predicting the Past platform?\nGoogle DeepMind\nPredicting the past with Ithaca — Google DeepMind\nCore: Ithaca deep neural network (restores ancient Greek inscriptions, identifies geographic origins, establishes creation dates). Architecture: Sparse multi-he...\nGitHub\nGitHub - google-deepmind/ithaca: Restoring and attributing ancient texts using deep neural networks · GitHub\nOverview: Ithaca is a deep neural network developed by DeepMind and academic partners for restoring, geographically attributing, and chronologically dating anci...\nPredicting the Past\nPredicting the Past\nContextualising, restoring, and attributing ancient texts Ithaca and Aeneas are generative AI models designed to assist historians in the study of ancient Greek...\nNature\nRestoring and attributing ancient texts using deep neural networks | Nature\nModel: Ithaca, a deep neural network trained on ancient Greek inscriptions (7th century BC to 5th century AD) from the I.PHI dataset (78,608 texts), simultaneou...",
 "transcript_raw": "Ithaca is a deep neural network developed by Google DeepMind and academic partners to restore, geographically attribute, and chronologically date damaged ancient Greek inscriptions. [[1](https://www.google.com/goto?url=CAESYQHrOzAV9bhPXwiPFAkk0WrtjwF69osLlHBUWNzzI69S1WtwZTqUCsUsFHJRmbHqvOTqawUMo9x8Q-JEjDS2uEGe-nwFB8K7bhK_9tMlb9RUSeY2GDejKtZrzQPcaK1YQ8A)]\nKey Capabilities\n   * Text Restoration: Restores missing or damaged characters with 62% accuracy on its own (raising human historian accuracy from 25% to 72% when collaborating). [[1](https://www.google.com/goto?url=CAESYQHrOzAV9bhPXwiPFAkk0WrtjwF69osLlHBUWNzzI69S1WtwZTqUCsUsFHJRmbHqvOTqawUMo9x8Q-JEjDS2uEGe-nwFB8K7bhK_9tMlb9RUSeY2GDejKtZrzQPcaK1YQ8A)]\n   * Geographical Attribution: Identifies the original geographic location across 84 regions with 71% accuracy. [[1](https://www.google.com/goto?url=CAESdQHrOzAV_0MWkpVTpN7uvNvXlOBaZ40iii-iOTMqzp09h7b0HxKyB-_B3anpWxAEboASgFYZPv2f1wL6NWVEbFQlf4ki9DS6MGXMox5jtuJ6YxNutsIntryMo_rhg8_NQ3HN7Mpl7kEKmaM9FC0kvaP9Eb726w), [2](https://www.google.com/goto?url=CAESagHrOzAVvUud25tCo9gLWQgmo4vGXjLWeVFxE5AU0Uki3Gg-oMDVXRFMODGH1hrVZkipV_2rQt9md-GkhXBTCM2McqHlXUfunJiUdmKCar8LAG0D4xDoHA6uzaN6yPDhb_m-D2SdWYGdhSk)]\n   * Chronological Dating: Dates inscriptions within 30 years of their true historical creation date. [[1](https://www.google.com/goto?url=CAESdQHrOzAV_0MWkpVTpN7uvNvXlOBaZ40iii-iOTMqzp09h7b0HxKyB-_B3anpWxAEboASgFYZPv2f1wL6NWVEbFQlf4ki9DS6MGXMox5jtuJ6YxNutsIntryMo_rhg8_NQ3HN7Mpl7kEKmaM9FC0kvaP9Eb726w)]\nHow It Works\n   * Architecture: Uses joint character and word embeddings with stacked transformer decoder blocks and sparse multi-head attention. [[1](https://www.google.com/goto?url=CAESagHrOzAVvUud25tCo9gLWQgmo4vGXjLWeVFxE5AU0Uki3Gg-oMDVXRFMODGH1hrVZkipV_2rQt9md-GkhXBTCM2McqHlXUfunJiUdmKCar8LAG0D4xDoHA6uzaN6yPDhb_m-D2SdWYGdhSk), [2](https://www.google.com/goto?url=CAESdQHrOzAV_0MWkpVTpN7uvNvXlOBaZ40iii-iOTMqzp09h7b0HxKyB-_B3anpWxAEboASgFYZPv2f1wL6NWVEbFQlf4ki9DS6MGXMox5jtuJ6YxNutsIntryMo_rhg8_NQ3HN7Mpl7kEKmaM9FC0kvaP9Eb726w)]\n   * Training Data: Trained on the Packard Humanities Institute (PHI) Searchable Greek Inscriptions dataset. [[1](https://www.google.com/goto?url=CAESdQHrOzAV_0MWkpVTpN7uvNvXlOBaZ40iii-iOTMqzp09h7b0HxKyB-_B3anpWxAEboASgFYZPv2f1wL6NWVEbFQlf4ki9DS6MGXMox5jtuJ6YxNutsIntryMo_rhg8_NQ3HN7Mpl7kEKmaM9FC0kvaP9Eb726w), [2](https://www.google.com/goto?url=CAESYQHrOzAV9bhPXwiPFAkk0WrtjwF69osLlHBUWNzzI69S1WtwZTqUCsUsFHJRmbHqvOTqawUMo9x8Q-JEjDS2uEGe-nwFB8K7bhK_9tMlb9RUSeY2GDejKtZrzQPcaK1YQ8A)]\n   * Access & Open Source: The code is open-sourced under the Apache-2.0 license via the [GitHub - Google-DeepMind Ithaca Repository](https://www.google.com/goto?url=CAESYQHrOzAV9bhPXwiPFAkk0WrtjwF69osLlHBUWNzzI69S1WtwZTqUCsUsFHJRmbHqvOTqawUMo9x8Q-JEjDS2uEGe-nwFB8K7bhK_9tMlb9RUSeY2GDejKtZrzQPcaK1YQ8A), and interactive research tools can be explored through [Predicting the Past](https://www.google.com/goto?url=CAESVgHrOzAV5xjfiCSPWOfGgOXOl09A2Ia9t8w6n3GOcZKu8sNgrR6XHIbDPTs6hdVLhRmsBR_nK-WQVer7zY6gPSmm1pIQT5soqR4Dqyo3prIvXmk6EpMN). [[1](https://www.google.com/goto?url=CAESVgHrOzAV5xjfiCSPWOfGgOXOl09A2Ia9t8w6n3GOcZKu8sNgrR6XHIbDPTs6hdVLhRmsBR_nK-WQVer7zY6gPSmm1pIQT5soqR4Dqyo3prIvXmk6EpMN), [2](https://www.google.com/goto?url=CAESYQHrOzAV9bhPXwiPFAkk0WrtjwF69osLlHBUWNzzI69S1WtwZTqUCsUsFHJRmbHqvOTqawUMo9x8Q-JEjDS2uEGe-nwFB8K7bhK_9tMlb9RUSeY2GDejKtZrzQPcaK1YQ8A)]\nWould you like to know more about how Ithaca connects with newer models like Aeneas or how historians use the Predicting the Past platform?\nGoogle DeepMind\nPredicting the past with Ithaca — Google DeepMind\nCore: Ithaca deep neural network (restores ancient Greek inscriptions, identifies geographic origins, establishes creation dates). Architecture: Sparse multi-he...\nGitHub\nGitHub - google-deepmind/ithaca: Restoring and attributing ancient texts using deep neural networks · GitHub\nOverview: Ithaca is a deep neural network developed by DeepMind and academic partners for restoring, geographically attributing, and chronologically dating anci...\nPredicting the Past\nPredicting the Past\nContextualising, restoring, and attributing ancient texts Ithaca and Aeneas are generative AI models designed to assist historians in the study of ancient Greek...\nNature\nRestoring and attributing ancient texts using deep neural networks | Nature\nModel: Ithaca, a deep neural network trained on ancient Greek inscriptions (7th century BC to 5th century AD) from the I.PHI dataset (78,608 texts), simultaneou...",
 "transcript_class": "CAPTURE-TIME VERBATIM RECORD — paste of the Overview expanded, with its source cards",
 "transcript_complete": "complete as supplied",
 "transcript_read": "READ IN FULL 2026-10-31",
 "per": 0.5,
 "per_corrected": null,
 "per_v": {
  "author": false,
  "inst": true,
  "id": false,
  "src": true
 },
 "per_note": "Google DeepMind named; no person; no identifier; DeepMind's page, the code and the Nature paper cited.",
 "availability": null,
 "per_dim": {
  "spec": "PER-dim rubric v1.1",
  "entity_type": "protocol",
  "target": {
   "ent": "TGT-rf-ithaca",
   "name": "Ithaca"
  },
  "qualifier": null,
  "reached": "same",
  "gate": "scored",
  "atoms": [
   {
    "atom": "name",
    "dim": "M",
    "weight": 3,
    "kind": "token",
    "present": true,
    "evidence": "Ithaca is a deep neural network"
   },
   {
    "atom": "producer",
    "dim": "M",
    "weight": 3,
    "kind": "role",
    "present": true,
    "evidence": "developed by Google DeepMind and academic partners"
   },
   {
    "atom": "claim boundary",
    "dim": "M",
    "weight": 2,
    "kind": "reading",
    "present": true,
    "evidence": "Restores missing or damaged characters with 62% accuracy on its own (raising human historian accuracy from 25% to 72% when collaborating)."
   },
   {
    "atom": "institution",
    "dim": "C",
    "weight": 2,
    "kind": "token",
    "present": false,
    "evidence": null
   },
   {
    "atom": "purpose",
    "dim": "D",
    "weight": 3,
    "kind": "reading",
    "present": true,
    "evidence": "to restore, geographically attribute, and chronologically date damaged ancient Greek inscriptions"
   },
   {
    "atom": "commitments",
    "dim": "D",
    "weight": 2,
    "kind": "reading",
    "present": true,
    "evidence": "The code is open-sourced under the Apache-2.0 license via the"
   },
   {
    "atom": "lineage",
    "dim": "D",
    "weight": 2,
    "kind": "role",
    "present": true,
    "evidence": "Trained on the Packard Humanities Institute (PHI) Searchable Greek Inscriptions dataset."
   },
   {
    "atom": "identifier",
    "dim": "D",
    "weight": 1,
    "kind": "role",
    "present": true,
    "evidence": "GitHub - Google-DeepMind Ithaca Repository"
   },
   {
    "atom": "deposit chain",
    "dim": "D",
    "weight": 1,
    "kind": "role",
    "present": true,
    "evidence": "interactive research tools can be explored through"
   }
  ],
  "inversions": [],
  "standing_substitutions": [],
  "scores": {
   "M": 0.0,
   "C": 1.0,
   "D": 0.0,
   "PER": 0.105
  },
  "src": true,
  "tombstone_pointer": false,
  "after_challenge": null,
  "basis": "reached: Ithaca is a deep neural network developed by Google DeepMind and academic partners to restore, geographically attribute, and chronologically date damaged ancient Greek inscriptions.; The operator's disambiguator 'DeepMind' selects the model from the homonym Ithaca (the island, the city); the composition is of the model throughout, by its own three tasks.",
  "read": {
   "by": "intake reader R2, spans and record quotes checked verbatim by scripts/per_backlog.py",
   "on": "2026-10-31",
   "brief": "rebuild/capture-registry/intake-20261031-rf-census/per/BRIEF.md",
   "external": true,
   "notes": "Claim boundary present: the three tasks, 62% alone and 25% to 72% with historians, 71% over 84 regions, and dating within 30 years. One departure noted, not charged: 'within 30 years of their true historical creation date' inflates the record's 'ground-truth ranges' (the epigraphers' proposed dates, L84, L191); the record's commitment that restorations are plausible suggestions, never certain, is not carried. Purpose present: the tasks are bound to the model and its assistive frame survives in 'when collaborating'. Commitments present: the open-sourcing under Apache-2.0 is carried as the model's own (the CC BY-SA weights and the public interface's purpose are not). Lineage: PHI named and bound as its training source; Pythia absent. Identifier present by the composition's own text naming the GitHub repository as the way to the code (link text, not a URL or card; flag for the operator if anchor text should not count). Deposit chain present: the interface's current home, Predicting the Past, is offered as the way to the work, and the closing offer ties Ithaca to 'newer models like Aeneas' (the record: upgraded in July 2025 to be powered by Aeneas). Institution absent: the 'academic partners' are not named."
  }
 },
 "per_dim_pending": null,
 "per_dim_v1_0": null,
 "sf": null,
 "sf_derived": null,
 "reading": "The innocent reading first. The Overview gives the model, its three tasks with their figures, how it is built and where it can be used, from its maker, its code and its paper.",
 "analysis": "A model is composed by its benchmark, accurately. What the field takes from Ithaca is the scale of the error: a machine dates an inscription at an average distance of 29.3 years from its bracket, with a median of 3 (the paper, card 4), near Enoch's mean error for the scrolls.",
 "d": "ITHACA, BY THE OPERATOR'S ADDRESS 'ITHACA DEEPMIND', ON GOOGLE'S AI OVERVIEW, SIGNED OUT, 2026-10-31: 'a deep neural network developed by Google DeepMind and academic partners to restore, geographically attribute, and chronologically date damaged ancient Greek inscriptions.' Capabilities with their figures, architecture, the PHI training data, the Apache-2.0 code, the Predicting the Past interface, an offer on Aeneas. Four cards: DeepMind, GitHub, Predicting the Past, Nature.",
 "d_full": "ITHACA, BY THE OPERATOR'S ADDRESS 'ITHACA DEEPMIND', ON GOOGLE'S AI OVERVIEW, SIGNED OUT, 2026-10-31: 'a deep neural network developed by Google DeepMind and academic partners to restore, geographically attribute, and chronologically date damaged ancient Greek inscriptions.' Capabilities with their figures, architecture, the PHI training data, the Apache-2.0 code, the Predicting the Past interface, an offer on Aeneas. Four cards: DeepMind, GitHub, Predicting the Past, Nature.",
 "d_truncated": false,
 "links": [
  {
   "url": "https://www.alexanarch.org/captures/ithaca-deepmind-aio-20261031/",
   "authority": "canonical",
   "note": "the capture's own record page; cite this form"
  },
  {
   "url": "https://www.alexanarch.org/captures/#ithaca-deepmind-aio-20261031",
   "authority": "gallery",
   "note": "the canonical gallery, anchored by slug"
  },
  {
   "url": "https://www.godkinggoogle.com/captures/#ithaca-deepmind-aio-20261031",
   "authority": "mirror",
   "note": "a window that renders from the archive's registry; may lag a deploy"
  },
  {
   "url": "https://www.leesharks.com/captures/#ithaca-deepmind-aio-20261031",
   "authority": "mirror",
   "note": "a window that renders from the archive's registry; may lag a deploy"
  },
  {
   "url": "https://www.machinemediation.org/captures/#ithaca-deepmind-aio-20261031",
   "authority": "mirror",
   "note": "a window that renders from the archive's registry; may lag a deploy"
  }
 ],
 "imgs": [],
 "img_urls": [],
 "cite": "https://www.alexanarch.org/captures/ithaca-deepmind-aio-20261031/",
 "citable_unit": "address — the exact issued string on one surface, per the Surface Rule (MANUS, 2026-08-15)",
 "addr_id": "ADDR-665778bf9317",
 "obs_id": "OBS-5e8377899856",
 "n_observations": 1,
 "observations": [],
 "dates": [
  "2026-10-31"
 ],
 "defects": [],
 "findings": [
  "THE MODEL HELD. 'Restores missing or damaged characters with 62% accuracy on its own (raising human historian accuracy from 25% to 72% when collaborating)', 'with 71% accuracy', 'within 30 years of their true historical creation date': the record's figures, from the Nature abstract. Checked 2026-10-31 against the Revelation First census record of 2026-10-30 and the sources it saved.",
  "THE STATED LIMIT AS FIGURES. The paper's own frame, 'Ithaca is designed to assist and expand the historian’s workflow', is carried only by the collaboration figures. The redating of key texts of Classical Athens, the record's second finding, is absent.",
  "FIGURES BEYOND THE RECORD. 'across 84 regions', the architecture ('stacked transformer decoder blocks and sparse multi-head attention'), the PHI dataset and 'the Apache-2.0 license' are in the Nature paper the census saved (card 4), which the record's fields did not carry; the licence is the paper's ('under Apache License 2.0').",
  "THE DATE OF THE MODEL ABSENT. The paper is of 2022; the composition gives no year."
 ],
 "series": null,
 "other_slugs": null,
 "collisions": null,
 "oq": null,
 "rounds": null,
 "turns": 1,
 "rerun": null,
 "rerun_alt": null,
 "heteronym": null,
 "model_attribution": null,
 "operator_disclosure": null,
 "longitudinal_priors": null,
 "longitudinal_successors": null,
 "related_deposits": [],
 "intended": null,
 "originator": {
  "name": "DeepMind with historians and collaborators; described in Nature (2022)",
  "relation": "external",
  "entity_type": "protocol",
  "spxi_treatment": "partial",
  "basis": "Not the archive's: Ithaca; its operator, as the record gives it: DeepMind with historians and collaborators; described in Nature (2022). Formalized by the Revelation First meta-aggregator record of 2026-10-30 (https://www.revelationfirst.com/aggregators/ithaca/), editor Lee Sharks; the record's fields carry verbatim quotes from the entity's own pages, saved with sha256."
 },
 "notes": {
  "prompts": "One query. Sent by the operator on 2026-10-31 (00:04 EDT) with nine others, one per entity of the Revelation First meta-aggregator, from the addresses the archive supplied for capture on 2026-10-30; eight pasted in one message with '$' between them and two attached as files (the Vesuvius Challenge attached twice, the copies byte-identical). Saved verbatim from the session transcript (rebuild/capture-registry/intake-20261031-rf-census/)."
 },
 "notice": null,
 "axn_root": "AXN:0B36.CAPTURE",
 "axn": "AXN:0B36.CAPTURE.🗿🪨🕊️🔃➖🌳",
 "record_url": "https://www.alexanarch.org/captures/ithaca-deepmind-aio-20261031/"
}