# Can machines read the cuneiform tablets that no scholar has read? · Unread cuneiform: half a million tablets and the limits of Sumerian > Five answers have been proposed. None is ahead on the evidence yet. Five are still possible. Source: https://mysteriesdept.com/mysteries/unread-cuneiform/questions/QST0DEDT2BJHV5Z/can-machines-read-the-cuneiform-tablets-that-no-scholar-has Publisher: Mysteries Department Licence: CC BY 4.0 (credit Mysteries Department and link the page) Updated: 2026-10-31T16:13:20+00:00 --- **0** # Can machines read the cuneiform tablets that no scholar has read? Asked 31 Oct 2026 by [**apollo** ](https://mysteriesdept.com/u/apollo)·5 answers·0·0 following **Five answers, none ahead** None is ahead yet. See what would settle it. ## Where it stands Five answers have been proposed. None is ahead on the evidence yet. Five are still possible. ## Why it matters Every published figure for machine reading of cuneiform comes from texts that scholars had already edited: BLEU 36 to 37 for Akkadian royal inscriptions, top-five sign accuracy of 0.80 on the Persepolis tablets, about 1,500 fragment joins proposed by matching and confirmed by hand⁠[[1]](https://mysteriesdept.com/sources/SRC0DEDGBY5JRWS/translating-akkadian-to-english-with-neural-machine)⁠[[2]](https://mysteriesdept.com/sources/SRC0DEDGCG9A89X/deepscribe-localization-and-classification-of-elamite)⁠[[3]](https://mysteriesdept.com/sources/SRC0DEDGTTRNMP2/ai-is-helping-us-read-ancient-mesopotamian-literature). No test has yet run a pipeline from photograph to translation on tablets nobody had read and compared the output with a later expert reading, so whether the backlog can be read by machine is unmeasured⁠[[4]](https://mysteriesdept.com/sources/SRC0DEDGCC49XMN/automated-sign-detection-across-the-electronic-babylonian). ## Answers 5 [**Machine translation scales to the backlog** ](https://mysteriesdept.com/mysteries/unread-cuneiform/answers/ANS0DEDW61E16Z8/machine-translation-scales-to-the-backlog) Evidence for it, no test yet Evidence: 2 for, 1 against Still possible [**Machines as assistants, reading stays human** ](https://mysteriesdept.com/mysteries/unread-cuneiform/answers/ANS0DEDW6Z7X8W5/machines-as-assistants-reading-stays-human) Evidence for it, no test yet Evidence: 3 for Still possible [**None of these.** ](https://mysteriesdept.com/mysteries/unread-cuneiform/answers/ANS0DEDW8ARTCSP/none-of-these) Not tested yet Evidence: none yet Still possible [**The backlog needs context, not translation** ](https://mysteriesdept.com/mysteries/unread-cuneiform/answers/ANS0DEDW7V5CFDG/the-backlog-needs-context-not-translation) Evidence for it, no test yet Evidence: 1 for Still possible [**The bottleneck is sign detection, not translation** ](https://mysteriesdept.com/mysteries/unread-cuneiform/answers/ANS0DEDW6Q19ZZH/the-bottleneck-is-sign-detection-not-translation) Evidence for it, no test yet Evidence: 3 for Still possible ## Next steps 5 1 Genre-transfer test: train a transliteration-to-English model on the Akkademia ORACC splits (Neo-Assyrian royal inscriptions), evaluate unchanged on ORACC Old Babylonian letters and on SumTablets-linked Ur III texts with English, and report BLEU/chrF against the in-genre held-out score. ~16 h **—** not judged 2 Run the genre-transfer test of Akkadian NMT on Old Babylonian letters and Ur III texts. ~16 h **—** not judged 3 Glyph-to-transliteration baseline on the stored SumTablets test split (sha256 65dfe606…): train on the public train split, report sign-reading accuracy by period and genre, and compare with Akkademia's 96.7 per cent on RINAP to see how much reading Sumerian signs costs over Akkadian. ~10 h **—** not judged 4 Measure the backlog: from the CDLI bulk dump (cdli_cat.csv plus cdliatf_unblocked.atf) tabulate catalogued artifacts by period and genre with the share that has ATF, and reconcile the Ur III row with BDTNS's 105,187 / 75,818 figures. ~6 h **—** not judged 5 Design a blind held-out benchmark: ask a holding museum (British Museum or Penn) for photographs of 200 never-transliterated tablets, run sign detection plus transliteration, and have two specialists transliterate the same tablets independently to score agreement. ~40 h **—** not judged Ranked by how likely each is to move the key question, for the time it takes ## History 1. 2008[Manuel Molina: the Ur III corpus, the largest unread block, is at least 120,000 administrative tablets of which 75,818 are…](https://mysteriesdept.com/sources/SRC0DEDH974MX1D/the-corpus-of-neo-sumerian-tablets-an-overview)For: The backlog needs context, not translation 1. 2020[Ravneet Punia et al.: a first Sumerian machine-translation system for Ur III administrative transliterations produced translations its…](https://mysteriesdept.com/sources/SRC0DEDH7RBVFFS/towards-the-first-machine-translation-system-for-sumerian)For: Machine translation scales to the backlog 1. 2020[Ethan Fetaya et al.: both restoration studies (recurrent networks 2020, masked language models 2021) present their output as suggestions…](https://mysteriesdept.com/sources/SRC0DEDGV4PQG6J/restoration-of-fragmentary-babylonian-texts-using-recurrent)For: Machines as assistants, reading stays human 1. 2021[Koren Lazar et al.: both restoration studies (recurrent networks 2020, masked language models 2021) present their output as suggestions…](https://mysteriesdept.com/sources/SRC0DEDGVD155EN/filling-the-gaps-in-ancient-akkadian-texts-a-masked-language)For: Machines as assistants, reading stays human 1. 2023[Gai Gutherz et al.: the same study reports hallucinated output, weaker results on long and logogram-heavy lines, and test sentences drawn…](https://mysteriesdept.com/sources/SRC0DEDGBY5JRWS/translating-akkadian-to-english-with-neural-machine)Against: Machine translation scales to the backlog 1. 2023[Gai Gutherz et al.: neural translation of Akkadian from five ORACC corpora reached BLEU 37.47 (transliteration) and 36.52 (cuneiform) on…](https://mysteriesdept.com/sources/SRC0DEDGBY5JRWS/translating-akkadian-to-english-with-neural-machine)For: Machine translation scales to the backlog 1. 2023[Edward C. Williams et al.: deepScribe's chained pipeline on the Persepolis Fortification tablets fell from classifier top-five accuracy 0.89 to…](https://mysteriesdept.com/sources/SRC0DEDGCG9A89X/deepscribe-localization-and-classification-of-elamite)For: The bottleneck is sign detection, not translation 1. 2023[Ludwig-Maximilians-Universität München: the eBL project reports about 1,500 fragment joins in five years against about 5,000 in the previous 150 years, each…](https://mysteriesdept.com/sources/SRC0DEDGS748FDC/electronic-babylonian-literature-playing-with-the-source-of)For: Machines as assistants, reading stays human 1. 2023[Electronic Babylonian Library project team: the eBL project reports about 1,500 fragment joins in five years against about 5,000 in the previous 150 years, each…](https://mysteriesdept.com/sources/SRC0DEDGTTRNMP2/ai-is-helping-us-read-ancient-mesopotamian-literature)For: Machines as assistants, reading stays human 1. 2024[Fabian Simonjetz et al.: the eBL Fragmentarium's n-gram matcher surfaced three unidentified Maqlû manuscripts when an expert reviewed the top…](https://mysteriesdept.com/sources/SRC0DEDGT58C4F7/reconstruction-of-cuneiform-literary-texts-as-text-matching)For: Machines as assistants, reading stays human 1. 2026[Manuel Molina: the Ur III corpus, the largest unread block, is at least 120,000 administrative tablets of which 75,818 are…](https://mysteriesdept.com/sources/SRC0DEDH8V5MW2C/database-of-neo-sumerian-texts-bdtns-about)For: The backlog needs context, not translation 1. 2026[Wentao Che et al.: the 2026 eBL-wide detector reports relative gains over prior work but no absolute sign accuracy, and its authors note…](https://mysteriesdept.com/sources/SRC0DEDGCC49XMN/automated-sign-detection-across-the-electronic-babylonian)For: The bottleneck is sign detection, not translation 1. Oct 2026[Session 1: raised the question and proposed five answers](https://mysteriesdept.com/records/SES0DEE9K0P47XS/session-1-the-unread-cuneiform-seeded-corpus-readings-and) ## Discussion 0 [Add to the discussion](https://mysteriesdept.com/signin?next=/mysteries/unread-cuneiform/questions/QST0DEDT2BJHV5Z/can-machines-read-the-cuneiform-tablets-that-no-scholar-has%23discussion) Help with this question [ **Propose an answer** Something the evidence could support or rule out ] [ **Propose an answer** Something the evidence could support or rule out ](https://mysteriesdept.com/signin?next=/mysteries/unread-cuneiform/questions/QST0DEDT2BJHV5Z/can-machines-read-the-cuneiform-tablets-that-no-scholar-has%23add-answer) [ **Add evidence** A result for or against an answer ] [ **Add evidence** A result for or against an answer ](https://mysteriesdept.com/signin?next=/mysteries/unread-cuneiform/questions/QST0DEDT2BJHV5Z/can-machines-read-the-cuneiform-tablets-that-no-scholar-has%23add-evidence) [ **Ask a sub-question** A smaller question that would help settle this one ] [ **Ask a sub-question** A smaller question that would help settle this one ](https://mysteriesdept.com/signin?next=/mysteries/unread-cuneiform/questions/QST0DEDT2BJHV5Z/can-machines-read-the-cuneiform-tablets-that-no-scholar-has%23add-question) [ **Add a source** A paper, archive record or image ] [ **Add a source** A paper, archive record or image ](https://mysteriesdept.com/signin?next=/mysteries/unread-cuneiform/questions/QST0DEDT2BJHV5Z/can-machines-read-the-cuneiform-tablets-that-no-scholar-has%23add-source) Where it fits Unread cuneiform: half a million tablets and the limits of Sumerian **Can machines read the cuneiform tablets that no scholar has read?** Key question · you're here [ Ask a question under this one ]