Neural Decipherment via Minimum-Cost Flow: from Ugaritic to Linear B · HackerPapers discussions

Neural Decipherment via Minimum-Cost Flow: from Ugaritic to Linear B — mentioned by mkw5053

Tue, 19 Aug 2025 04:48:39 GMT

Very cool! I was just going down a rabbit hole yesterday about the use of AI techniques (or lack of success) in deciphering still-forgotten languages. Unsupervised models have partially cracked Ugaritic and Linear B [0], and Pythia/Ithaca restore Greek inscriptions at scale [1], but Linear A or Proto-Elamite still stall because the corpora are too small and there is no bilingual ‘Rosetta Stone’. The most promising direction now seems to be hybrid pipelines that combine vision encoders to normalize glyphs with constrained decoders guided by phonotactic priors. [0]…

Paper and discussion history. A mention is not an endorsement.

Neural Decipherment via Minimum-Cost Flow: from Ugaritic to Linear B — mentioned by changoplatanero

Sun, 15 May 2022 15:50:26 GMT

You mean like this? https://arxiv.org/pdf/1906.06718.pdf

Paper and discussion history. A mention is not an endorsement.

Neural Decipherment via Minimum-Cost Flow: from Ugaritic to Linear B — HN submission

Wed, 21 Oct 2020 15:12:05 GMT

Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B

30 points; 13 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Neural Decipherment via Minimum-Cost Flow: from Ugaritic to Linear B — HN submission

Mon, 22 Jul 2019 23:16:25 GMT

Neural Decipherment via Minimum-Cost Flow: from Ugaritic to Linear B

76 points; 7 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Neural Decipherment via Minimum-Cost Flow: from Ugaritic to Linear B — HN submission

Mon, 15 Jul 2019 11:28:02 GMT

Neural Decipherment via Minimum-Cost Flow: From Ugaritic to Linear B

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.