{ "id": "AIBD-0277", "title": "Neural machine translation renders Akkadian cuneiform into English at BLEU4 36.5–37.5", "summary": "Tel Aviv and Ariel University researchers trained CNN-based models translating Akkadian from cuneiform glyphs or transliteration to English, best on short sentences; a capability demonstration only.", "date_observed": "2023-05-02", "date_basis": "unassessed", "date_precision": "unassessed", "benefit_status": null, "source_collection": null, "source_entry": null, "account": "Gutherz, Gordin, Saenz, Levy and Berant (PNAS Nexus, 2 May 2023) trained neural machine translation models to render Akkadian into English. One works directly from cuneiform Unicode glyphs (C2E), one from transliteration (T2E). They report BLEU4 scores of 36.52 and 37.47, improvements of 9.43 and 13.96 over a translation-memory baseline. Results were best on sentences of about 118 characters or fewer. They propose a human-in-the-loop system in which scholars correct outputs. No passage read and accepted by scholars is named. The authors report hallucinations in longer or fragmented texts. The work stands as a capability demonstration. The group's 2020 PLoS ONE paper (Akkademia) reported up to 97% accuracy for automatic transliteration and segmentation.", "category": "general benefit", "unexpected": false, "purpose": "Automatically translate Akkadian cuneiform texts into English to assist scholars", "system": "Akkadian-to-English neural machine translation (C2E and T2E models)", "model_version": "Fairseq convolutional model", "developer": "Tel Aviv University (Gutherz, Levy, Berant) and Ariel University (Gordin)", "deployer": "None (research demonstration)", "location": "Israel", "domain": "Digital humanities; Assyriology", "segment": "Assyriologists and interested lay readers", "benefit_type": "Improved Task Accuracy", "taxonomy_note": null, "registry_ids": {}, "ratings": [ { "group": "personal", "depth": null, "reason": "No individual beneficiary." }, { "group": "organizational", "depth": null, "reason": "No deployment reported." }, { "group": "societal", "depth": null, "reason": "No attributable event." } ], "ai_role": { "status": "assessed", "statement": "The models' translation quality is reported by their developers in a peer-reviewed paper. No scholar-accepted translation of a specific text is attributed to them. The authors note limits on longer and fragmented texts." }, "evidence": [ { "kind": "peer-reviewed study", "citation": "Gutherz G, Gordin S, Saenz L, Levy O, Berant J. PNAS Nexus 2(5):pgad096 (2023)", "title": "Translating Akkadian to English with neural machine translation", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10153418/", "date": "2023-05-02", "written_by": "makers", "quote": "we get 36.52 and 37.47 Best Bilingual Evaluation Understudy 4 (BLEU4) scores for C2E and T2E, respectively" }, { "kind": "peer-reviewed study", "citation": "Gordin S, Gutherz G, Elazary A, et al. PLoS ONE 15(10):e0240511 (2020)", "title": "Reading Akkadian cuneiform using natural language processing", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7592802/", "date": "2020-10-28", "written_by": "makers", "quote": "with an accuracy reaching up to 97% for automatically transliterating and segmenting standard Unicode cuneiform glyphs into words." }, { "kind": "preprint", "citation": "Wang Z (14 May 2026). TabletCraft: Bridging a 4,000-Year Cultural Gap with Bidirectional Akkadian NMT and Cuneiform Rendering. arXiv:2608.02609.", "title": "TabletCraft: Bridging a 4,000-Year Cultural Gap with Bidirectional Akkadian NMT and Cuneiform Rendering", "url": "https://arxiv.org/abs/2608.02609", "date": "2026-05-14", "written_by": "independent", "quote": null } ], "replication": { "produced_by": [ "Convolutional neural network translation models trained on digitized Akkadian-English pairs" ], "must_hold": [ "Human scholars correct outputs, especially for long or fragmentary texts" ], "echoes": [ { "date": "2020-10-28", "setting": "Earlier Akkademia transliteration and segmentation models by the same group", "model_version": null, "outcome": "recurred", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7592802/" } ] }, "paired_incident": null, "headline": "Tel Aviv and Ariel models translate Akkadian at BLEU4 36.5–37.5", "metric": "BLEU4 scores of 36.52 and 37.47", "comparator": { "against": "a translation memory baseline on the same test data", "before": "the translation memory baseline achieved a BLEU4 score of 23.51", "after": "36.52 and 37.47 Best Bilingual Evaluation Understudy 4 (BLEU4) scores for C2E and T2E, respectively", "basis": "measured", "quote": "we get 36.52 and 37.47 Best Bilingual Evaluation Understudy 4 (BLEU4) scores for C2E and T2E, respectively. For C2E, our model is better than the translation memory baseline in 9.43, and for T2E, the difference is even higher and stands at 13.96.", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10153418/" }, "later": { "status": "persisted", "date": "2026-05-14", "note": "The approach was built upon rather than contradicted: a May 2026 arXiv preprint by Zhaohui Wang (USC) cites the 2023 system and reports a bidirectional ByT5-based model reaching 49.1 BLEU for Akkadian-to-English on the same Akkademia validation split; no later report from the Tel Aviv and Ariel group itself was found.", "quote": "On the held-out Akkademia validation split (2,812 samples), we report 49.1 BLEU for Akkadian-to-English and 48.5 BLEU for English-to-Akkadian", "url": "https://arxiv.org/abs/2608.02609" }, "value": null, "developer_country": "Israel", "deployer_country": null, "unrated_because": null, "brief_notes": null, "brief": null, "revisions": [ { "date": "2026-10-20", "change": "Record drafted from the cited sources.", "source": null, "editor": "AIBD" }, { "date": "2026-10-21", "change": "Benefit type set to Improved Task Accuracy under Trustible's full taxonomy of 23 types; it was Improved Knowledge Discovery among the six first used.", "source": "https://trustible.ai/post/a-framework-for-measuring-the-benefits-of-ai/", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Added what the benefit is measured against, what happened since, the model version from the cited sources.", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10153418/; https://arxiv.org/abs/2608.02609; https://pmc.ncbi.nlm.nih.gov/articles/PMC10153418/", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Country of the developer added from the organization's own site or a registry.", "source": "https://en.wikipedia.org/wiki/Tel_Aviv_University", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Prose pass: account, the system's part rewritten in shorter sentences with abbreviations expanded; facts, figures and quotations unchanged.", "source": null, "editor": "AIBD" }, { "date": "2026-10-21", "change": "Notes about which sources could be read were reworded for readers; facts, figures and quotations unchanged.", "source": null, "editor": "AIBD" } ], "confidence": "reported", "shade": "medium", "claim_status": "Independently reported", "claim_basis": "Sources: one independent preprint and two peer-reviewed studies from the makers. The cited evidence includes no independent study or report and no finding by a regulator or court.", "profile": { "case_id": "AIBD-0277", "catalog_version": "1.0.0", "source_fingerprint": "48b8e8010ae6a7f1f4810f432a3896a49ed26393229a7169e44d10978d1d443b", "collection_refs": [ { "collection_id": "aibd", "version": "snapshot-8bf7cfec9cb185aa" }, { "collection_id": "aibd", "version": "2026-10-25" } ], "source_ids": [ "AIBD-0277-source-0001", "AIBD-0277-source-0002", "AIBD-0277-source-0003" ], "families": { "strategic assessments": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "record_ids": [] }, "evaluations": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "record_ids": [] }, "monetary observations": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "record_ids": [] }, "institutional decisions": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "record_ids": [] } }, "fields": { "actors[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "developer": "Tel Aviv University (Gutherz, Levy, Berant) and Ariel University (Gordin)", "deployer": "None (research demonstration)" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "developer", "deployer" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "actors[].ownership_control": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "ai_contribution": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "ai_role": { "status": "assessed", "statement": "The models' translation quality is reported by their developers in a peer-reviewed paper. No scholar-accepted translation of a specific text is attributed to them. The authors note limits on longer and fragmented texts." } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ai_role" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "ai_relationship": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "category": "general benefit", "ai_role": { "status": "assessed", "statement": "The models' translation quality is reported by their developers in a peer-reviewed paper. No scholar-accepted translation of a specific text is attributed to them. The authors note limits on longer and fragmented texts." } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ai_role", "category" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "assessments[].alternatives": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "assessments[].assessor_date": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "assessments[].depth": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "assessments[].display_confidence": { "state": "value_present", "value": { "kind": "structured", "data": { "confidence": "medium", "evidence_word": "reported", "claim_status": "Independently reported", "basis": "Sources: one independent preprint and two peer-reviewed studies from the makers. The cited evidence includes no independent study or report and no finding by a regulator or court.", "interpretation": "Existing display rule preserved; the richer outcome-specific confidence question still needs review." } }, "provenance": { "kind": "curatorial", "basis": "Derived from the existing case display properties; not a new evidentiary judgment.", "record_fields": [], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "assessments[].rationale": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "ratings": [ { "group": "personal", "reason": "No individual beneficiary." }, { "group": "organizational", "reason": "No deployment reported." }, { "group": "societal", "reason": "No attributable event." } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ratings" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "assessments[].rubric_version": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "assessments[].status_basis": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "assessments[].target": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "ratings": [ { "group": "personal", "reason": "No individual beneficiary." }, { "group": "organizational", "reason": "No deployment reported." }, { "group": "societal", "reason": "No attributable event." } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ratings" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "availability": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "brief": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "case_kind": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "case_links[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "claims[].evidence_links": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "claims[].evidence_strength": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "claims[].missingness": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "claims[].proposition": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "cross_border_flows[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "decisions[].action_outcome": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "institutional decisions" }, "decisions[].grounds": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "institutional decisions" }, "decisions[].institution_task": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "institutional decisions" }, "decisions[].success_criterion": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "institutional decisions" }, "deployment_stage": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "economics[].amount": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].costs[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].counterfactual": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].currency_price_basis": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].gross_net": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].measure_type": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].period_unit": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].realization": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].recipient_payer": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].return_payback": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "economics[].valuation_method": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "monetary observations" }, "evaluations[].ai_uplift": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].comparator": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].design": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].effect": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].exposure": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].id": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].measurement_method": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].metric": { "state": "not_assessed", "provenance": { "kind": "legacy_record", "basis": "Mapped from the existing case description without a new external source review.", "record_fields": [ "metric" ], "source_refs": [] }, "review_state": "not_assessed", "coverage": "partial", "family": "evaluations", "display": { "answer": "BLEU4 scores of 36.52 and 37.47", "editor_note": "" } }, "evaluations[].reproduction_materials": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].setting_sample": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "evaluations[].uncertainty_limitations": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "evaluations" }, "external_references[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "follow_up[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "later": { "status": "persisted", "date": "2026-05-14", "note": "The approach was built upon rather than contradicted: a May 2026 arXiv preprint by Zhaohui Wang (USC) cites the 2023 system and reports a bidirectional ByT5-based model reaching 49.1 BLEU for Akkadian-to-English on the same Akkademia validation split; no later report from the Tel Aviv and Ariel group itself was found.", "quote": "On the held-out Akkademia validation split (2,812 samples), we report 49.1 BLEU for Akkadian-to-English and 48.5 BLEU for English-to-Akkadian", "url": "https://arxiv.org/abs/2608.02609" } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "later" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "human_oversight": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "id": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "id": "AIBD-0277" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "id" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "infrastructure_dependencies[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].beneficiaries": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].beneficiary_count": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].benefit_type": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "benefit_type": "Improved Task Accuracy" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "benefit_type" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "outcomes[].claim": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "account": "Gutherz, Gordin, Saenz, Levy and Berant (PNAS Nexus, 2 May 2023) trained neural machine translation models to render Akkadian into English. One works directly from cuneiform Unicode glyphs (C2E), one from transliteration (T2E). They report BLEU4 scores of 36.52 and 37.47, improvements of 9.43 and 13.96 over a translation-memory baseline. Results were best on sentences of about 118 characters or fewer. They propose a human-in-the-loop system in which scholars correct outputs. No passage read and accepted by scholars is named. The authors report hallucinations in longer or fragmented texts. The work stands as a capability demonstration. The group's 2020 PLoS ONE paper (Akkademia) reported up to 97% accuracy for automatic transliteration and segmentation.", "benefit_type": "Improved Task Accuracy" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "account", "benefit_type" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "outcomes[].distribution": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].duration": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].id": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].stakeholder_group": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "ratings": [ { "group": "personal", "reason": "No individual beneficiary." }, { "group": "organizational", "reason": "No deployment reported." }, { "group": "societal", "reason": "No attributable event." } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "ratings" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "outcomes[].status": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].tradeoffs": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "outcomes[].unexpected": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "unexpected": false, "purpose": "Automatically translate Akkadian cuneiform texts into English to assist scholars" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "unexpected", "purpose" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "places[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "location": "Israel", "developer_country": "Israel" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "location", "developer_country" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "purpose": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "purpose": "Automatically translate Akkadian cuneiform texts into English to assist scholars" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "purpose" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "record_dates": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "date_observed": "2023-05-02", "revisions": [ { "date": "2026-10-20", "change": "Record drafted from the cited sources.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Benefit type set to Improved Task Accuracy under Trustible's full taxonomy of 23 types; it was Improved Knowledge Discovery among the six first used.", "source": "https://trustible.ai/post/a-framework-for-measuring-the-benefits-of-ai/", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Added what the benefit is measured against, what happened since, the model version from the cited sources.", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10153418/; https://arxiv.org/abs/2608.02609; https://pmc.ncbi.nlm.nih.gov/articles/PMC10153418/", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Country of the developer added from the organization's own site or a registry.", "source": "https://en.wikipedia.org/wiki/Tel_Aviv_University", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Prose pass: account, the system's part rewritten in shorter sentences with abbreviations expanded; facts, figures and quotations unchanged.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Notes about which sources could be read were reworded for readers; facts, figures and quotations unchanged.", "editor": "AIBD" } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "date_observed", "revisions" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "replication.attempts[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "replication": { "produced_by": [ "Convolutional neural network translation models trained on digitized Akkadian-English pairs" ], "must_hold": [ "Human scholars correct outputs, especially for long or fragmentary texts" ], "echoes": [ { "date": "2020-10-28", "setting": "Earlier Akkademia transliteration and segmentation models by the same group", "outcome": "recurred", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7592802/" } ] } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "replication" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "replication.conditions": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "replication": { "produced_by": [ "Convolutional neural network translation models trained on digitized Akkadian-English pairs" ], "must_hold": [ "Human scholars correct outputs, especially for long or fragmentary texts" ], "echoes": [ { "date": "2020-10-28", "setting": "Earlier Akkademia transliteration and segmentation models by the same group", "outcome": "recurred", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7592802/" } ] } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "replication" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "revision_history[]": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "revisions": [ { "date": "2026-10-20", "change": "Record drafted from the cited sources.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Benefit type set to Improved Task Accuracy under Trustible's full taxonomy of 23 types; it was Improved Knowledge Discovery among the six first used.", "source": "https://trustible.ai/post/a-framework-for-measuring-the-benefits-of-ai/", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Added what the benefit is measured against, what happened since, the model version from the cited sources.", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10153418/; https://arxiv.org/abs/2608.02609; https://pmc.ncbi.nlm.nih.gov/articles/PMC10153418/", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Country of the developer added from the organization's own site or a registry.", "source": "https://en.wikipedia.org/wiki/Tel_Aviv_University", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Prose pass: account, the system's part rewritten in shorter sentences with abbreviations expanded; facts, figures and quotations unchanged.", "editor": "AIBD" }, { "date": "2026-10-21", "change": "Notes about which sources could be read were reworded for readers; facts, figures and quotations unchanged.", "editor": "AIBD" } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "revisions" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "sector": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "domain": "Digital humanities; Assyriology", "segment": "Assyriologists and interested lay readers" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "domain", "segment" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "sources[].access_version": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "sources[].citation_locator": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "evidence": [ { "kind": "peer-reviewed study", "citation": "Gutherz G, Gordin S, Saenz L, Levy O, Berant J. PNAS Nexus 2(5):pgad096 (2023)", "title": "Translating Akkadian to English with neural machine translation", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10153418/", "date": "2023-05-02", "written_by": "makers", "quote": "we get 36.52 and 37.47 Best Bilingual Evaluation Understudy 4 (BLEU4) scores for C2E and T2E, respectively" }, { "kind": "peer-reviewed study", "citation": "Gordin S, Gutherz G, Elazary A, et al. PLoS ONE 15(10):e0240511 (2020)", "title": "Reading Akkadian cuneiform using natural language processing", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7592802/", "date": "2020-10-28", "written_by": "makers", "quote": "with an accuracy reaching up to 97% for automatically transliterating and segmenting standard Unicode cuneiform glyphs into words." }, { "kind": "preprint", "citation": "Wang Z (14 May 2026). TabletCraft: Bridging a 4,000-Year Cultural Gap with Bidirectional Akkadian NMT and Cuneiform Rendering. arXiv:2608.02609.", "title": "TabletCraft: Bridging a 4,000-Year Cultural Gap with Bidirectional Akkadian NMT and Cuneiform Rendering", "url": "https://arxiv.org/abs/2608.02609", "date": "2026-05-14", "written_by": "independent" } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "evidence" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "sources[].id": { "state": "value_present", "value": { "kind": "identifiers", "identifiers": [ "AIBD-0277-source-0001", "AIBD-0277-source-0002", "AIBD-0277-source-0003" ], "role": "Existing evidence items, in their original order" }, "provenance": { "kind": "curatorial", "basis": "Stable within the fingerprinted record version; no source content or source review has been added.", "record_fields": [], "source_refs": [] }, "review_state": "record_mapped", "coverage": "complete" }, "sources[].language": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "sources[].relationship": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "evidence": [ { "kind": "peer-reviewed study", "citation": "Gutherz G, Gordin S, Saenz L, Levy O, Berant J. PNAS Nexus 2(5):pgad096 (2023)", "title": "Translating Akkadian to English with neural machine translation", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10153418/", "date": "2023-05-02", "written_by": "makers", "quote": "we get 36.52 and 37.47 Best Bilingual Evaluation Understudy 4 (BLEU4) scores for C2E and T2E, respectively" }, { "kind": "peer-reviewed study", "citation": "Gordin S, Gutherz G, Elazary A, et al. PLoS ONE 15(10):e0240511 (2020)", "title": "Reading Akkadian cuneiform using natural language processing", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7592802/", "date": "2020-10-28", "written_by": "makers", "quote": "with an accuracy reaching up to 97% for automatically transliterating and segmenting standard Unicode cuneiform glyphs into words." }, { "kind": "preprint", "citation": "Wang Z (14 May 2026). TabletCraft: Bridging a 4,000-Year Cultural Gap with Bidirectional Akkadian NMT and Cuneiform Rendering. arXiv:2608.02609.", "title": "TabletCraft: Bridging a 4,000-Year Cultural Gap with Bidirectional Akkadian NMT and Cuneiform Rendering", "url": "https://arxiv.org/abs/2608.02609", "date": "2026-05-14", "written_by": "independent" } ] } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "evidence" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "sources[].upstream_sources": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "strategic_assessments[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial", "family": "strategic assessments" }, "summary": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "summary": "Tel Aviv and Ariel University researchers trained CNN-based models translating Akkadian from cuneiform glyphs or transliteration to English, best on short sentences; a capability demonstration only.", "account": "Gutherz, Gordin, Saenz, Levy and Berant (PNAS Nexus, 2 May 2023) trained neural machine translation models to render Akkadian into English. One works directly from cuneiform Unicode glyphs (C2E), one from transliteration (T2E). They report BLEU4 scores of 36.52 and 37.47, improvements of 9.43 and 13.96 over a translation-memory baseline. Results were best on sentences of about 118 characters or fewer. They propose a human-in-the-loop system in which scholars correct outputs. No passage read and accepted by scholars is named. The authors report hallucinations in longer or fragmented texts. The work stands as a capability demonstration. The group's 2020 PLoS ONE paper (Akkademia) reported up to 97% accuracy for automatic transliteration and segmentation." } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "summary", "account" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "systems[].autonomy_permissions": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "systems[].component_role": { "state": "value_present", "value": { "kind": "text", "text": "The models translated Akkadian text into English." }, "provenance": { "kind": "legacy_record", "basis": "Restated the mechanical task already described in the existing system and AI-role fields. No new source review is claimed.", "record_fields": [ "system", "ai_role" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "systems[].deployment_configuration": { "state": "value_present", "value": { "kind": "text", "text": "Researchers used a convolutional model implemented in Fairseq." }, "provenance": { "kind": "legacy_record", "basis": "Separated the model, software, hardware, or operation already described in the existing record. No new source review is claimed.", "record_fields": [ "model_version", "system", "ai_role" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "systems[].inputs_outputs": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "systems[].model_name": { "state": "value_present", "value": { "kind": "text", "text": "A convolutional model implemented in Fairseq." }, "provenance": { "kind": "legacy_record", "basis": "Separated the model, software, hardware, or operation already described in the existing record. No new source review is claimed.", "record_fields": [ "model_version", "system", "ai_role" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "systems[].model_version": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "model_version": "Fairseq convolutional model" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "model_version" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial", "display": { "answer": "Exact release not established.", "editor_note": "" } }, "systems[].product_name": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "system": "Akkadian-to-English neural machine translation (C2E and T2E models)" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "system" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "systems[].provider": { "state": "value_present", "value": { "kind": "structured", "data": { "application_developer": "Tel Aviv University (Gutherz, Levy, Berant) and Ariel University (Gordin)" } }, "provenance": { "kind": "legacy_record", "basis": "Mapped from the existing case description without a new external source review.", "record_fields": [ "developer" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "taxonomy_crosswalks[]": { "state": "not_assessed", "review_state": "not_assessed", "coverage": "partial" }, "title": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "title": "Neural machine translation renders Akkadian cuneiform into English at BLEU4 36.5–37.5", "headline": "Tel Aviv and Ariel models translate Akkadian at BLEU4 36.5–37.5" } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "title", "headline" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" }, "version": { "state": "value_present", "value": { "kind": "identifiers", "identifiers": [ "48b8e8010ae6a7f1f4810f432a3896a49ed26393229a7169e44d10978d1d443b" ], "role": "SHA-256 of normalized original case data" }, "provenance": { "kind": "curatorial", "basis": "Deterministic content version assigned by the local migration.", "record_fields": [], "source_refs": [] }, "review_state": "record_mapped", "coverage": "complete" }, "workflow": { "state": "value_present", "value": { "kind": "record_fields", "fields": { "account": "Gutherz, Gordin, Saenz, Levy and Berant (PNAS Nexus, 2 May 2023) trained neural machine translation models to render Akkadian into English. One works directly from cuneiform Unicode glyphs (C2E), one from transliteration (T2E). They report BLEU4 scores of 36.52 and 37.47, improvements of 9.43 and 13.96 over a translation-memory baseline. Results were best on sentences of about 118 characters or fewer. They propose a human-in-the-loop system in which scholars correct outputs. No passage read and accepted by scholars is named. The authors report hallucinations in longer or fragmented texts. The work stands as a capability demonstration. The group's 2020 PLoS ONE paper (Akkademia) reported up to 97% accuracy for automatic transliteration and segmentation.", "replication": { "produced_by": [ "Convolutional neural network translation models trained on digitized Akkadian-English pairs" ], "must_hold": [ "Human scholars correct outputs, especially for long or fragmentary texts" ], "echoes": [ { "date": "2020-10-28", "setting": "Earlier Akkademia transliteration and segmentation models by the same group", "outcome": "recurred", "source": "https://pmc.ncbi.nlm.nih.gov/articles/PMC7592802/" } ] } } }, "provenance": { "kind": "legacy_record", "basis": "Preserved from the existing AIBD record; external sources were not rechecked.", "record_fields": [ "account", "replication" ], "source_refs": [] }, "review_state": "record_mapped", "coverage": "partial" } }, "summary": { "total": 91, "addressed": 91, "known": 32, "reviewed": 0, "record_mapped": 32, "not_assessed": 59, "complete": false, "by_state": { "value_present": 32, "not_reported": 0, "unresolved": 0, "not_assessed": 59, "not_applicable": 0, "withheld": 0, "source_unavailable": 0 } } } }