@inproceedings{zhou-anderson-2026-transfers,
title = "What Transfers Across Cuneiform? Script-Level Substrate and Language-Specific Regularities via Segmentation and Representation",
author = "Zhou, Chuanjun and
Anderson, Adam G",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.2169/",
pages = "49324--49340",
ISBN = "979-8-89176-459-0",
abstract = "Cuneiform recorded typologically unrelated languages for more than three millennia, raising a natural question: which of its statistical regularities are properties of the shared writing system, and which belong to individual languages? We answer this with two probe pairs targeting the same script-versus-language decomposition. The first compares two unsupervised, dictionary-free segmentation methods: transitional probability (TP), grounded in Saffran et al. (1996), and Morfessor (Creutz and Lagus, 2007). To our knowledge, this is the first systematic evaluation of dictionary-free segmentation on cuneiform; the only prior published baseline (Homburg and Chiarcos, 2016) relies on a hand-built dictionary (MaxMatch, F1 = 0.74). Both our methods reach F1 $>$ 0.97 within language (Morfessor reaches F1 $>$ 0.99 on Akkadian and Sumerian), exceeding MaxMatch by more than 23 F1 points, but diverge sharply in cross-language transfer: TP holds F1 $>$ 0.95 across all six language pairs, while Morfessor collapses to F1 = 0.27 from the 13K-token Elamite source. A learning curve and a Hittite stress test (34.7{\%} foreign-script logograms, where TP drops 35 points and Morfessor only 8) further isolate the two underlying mechanisms. The second probe pair extends the same decomposition to representations: concatenating Unicode and Latin features improves classification in 11 of 12 morphosyntactic tasks ($p < 0.01$). We release cunei-tools and a three-language benchmark."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhou-anderson-2026-transfers">
<titleInfo>
<title>What Transfers Across Cuneiform? Script-Level Substrate and Language-Specific Regularities via Segmentation and Representation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chuanjun</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adam</namePart>
<namePart type="given">G</namePart>
<namePart type="family">Anderson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Cuneiform recorded typologically unrelated languages for more than three millennia, raising a natural question: which of its statistical regularities are properties of the shared writing system, and which belong to individual languages? We answer this with two probe pairs targeting the same script-versus-language decomposition. The first compares two unsupervised, dictionary-free segmentation methods: transitional probability (TP), grounded in Saffran et al. (1996), and Morfessor (Creutz and Lagus, 2007). To our knowledge, this is the first systematic evaluation of dictionary-free segmentation on cuneiform; the only prior published baseline (Homburg and Chiarcos, 2016) relies on a hand-built dictionary (MaxMatch, F1 = 0.74). Both our methods reach F1 > 0.97 within language (Morfessor reaches F1 > 0.99 on Akkadian and Sumerian), exceeding MaxMatch by more than 23 F1 points, but diverge sharply in cross-language transfer: TP holds F1 > 0.95 across all six language pairs, while Morfessor collapses to F1 = 0.27 from the 13K-token Elamite source. A learning curve and a Hittite stress test (34.7% foreign-script logograms, where TP drops 35 points and Morfessor only 8) further isolate the two underlying mechanisms. The second probe pair extends the same decomposition to representations: concatenating Unicode and Latin features improves classification in 11 of 12 morphosyntactic tasks (p < 0.01). We release cunei-tools and a three-language benchmark.</abstract>
<identifier type="citekey">zhou-anderson-2026-transfers</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2169/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>49324</start>
<end>49340</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T What Transfers Across Cuneiform? Script-Level Substrate and Language-Specific Regularities via Segmentation and Representation
%A Zhou, Chuanjun
%A Anderson, Adam G.
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F zhou-anderson-2026-transfers
%X Cuneiform recorded typologically unrelated languages for more than three millennia, raising a natural question: which of its statistical regularities are properties of the shared writing system, and which belong to individual languages? We answer this with two probe pairs targeting the same script-versus-language decomposition. The first compares two unsupervised, dictionary-free segmentation methods: transitional probability (TP), grounded in Saffran et al. (1996), and Morfessor (Creutz and Lagus, 2007). To our knowledge, this is the first systematic evaluation of dictionary-free segmentation on cuneiform; the only prior published baseline (Homburg and Chiarcos, 2016) relies on a hand-built dictionary (MaxMatch, F1 = 0.74). Both our methods reach F1 > 0.97 within language (Morfessor reaches F1 > 0.99 on Akkadian and Sumerian), exceeding MaxMatch by more than 23 F1 points, but diverge sharply in cross-language transfer: TP holds F1 > 0.95 across all six language pairs, while Morfessor collapses to F1 = 0.27 from the 13K-token Elamite source. A learning curve and a Hittite stress test (34.7% foreign-script logograms, where TP drops 35 points and Morfessor only 8) further isolate the two underlying mechanisms. The second probe pair extends the same decomposition to representations: concatenating Unicode and Latin features improves classification in 11 of 12 morphosyntactic tasks (p < 0.01). We release cunei-tools and a three-language benchmark.
%U https://aclanthology.org/2026.emnlp-main.2169/
%P 49324-49340
Markdown (Informal)
[What Transfers Across Cuneiform? Script-Level Substrate and Language-Specific Regularities via Segmentation and Representation](https://aclanthology.org/2026.emnlp-main.2169/) (Zhou & Anderson, EMNLP 2026)
ACL