Pith. sign in

REVIEW 3 major objections 6 minor 41 references

Ab Antiquo: Neural Proto-language Reconstruction

T0 review · 3 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A character-level neural encoder-decoder trained on Romance cognates reconstructs unseen Latin proto-words, achieving 64.1% exact reconstruction and average edit distance 0.65 on the orthographic test set.

desk verdict A solid, honest empirical paper that sets a new benchmark for Romance proto-word reconstruction; the headline numbers are plausible but rest on a partially released dataset and no variance reporting. read the letter →

arxiv 1908.02477 v3 pith:DSAUU4GS submitted 2019-08-07 cs.CL

Carlo Meloni , Shauli Ravfogel , Yoav Goldberg This is my paper · ORCID
classification cs.CL
keywords proto-languagereconstructioncomparativemethodRomancelanguagescharacter-levelneuralnetworkshistoricallinguisticssoundchangecognatesattentionmechanism
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper asks whether the comparative method of historical linguistics—reconstructing an ancestral proto-word from its surviving cognates—can be automated by a character-level neural network. It introduces a dataset of 8,799 Latin-to-Romance comparative entries, with cognates in French, Italian, Spanish, Portuguese and Romanian, in both written and IPA-transcribed forms. On unseen test cognates, the network reconstructs the exact Latin word 64.1% of the time in the orthographic setting, with average edit distance 0.65, improving on earlier reported results; phonetic-input performance is lower mainly because Latin vowel-length distinctions are hard to recover. Error analysis shows the model's mistakes cluster around well-documented sound changes such as high-mid vowel alternations, segment deletion, cluster simplification and morphological regularization, and a synthetic rule test confirms it internalizes many systematic phonological shifts. If this stands, automatic reconstruction can support historical linguists by scaling up proto-lexicon recovery and by highlighting which sound changes are genuinely opaque from daughter languages alone.

What carries the argument

The model is a character-level encoder-decoder with attention, built from GRU networks of 150 cells. The encoder processes the cognate set as a sequence of characters from all daughter languages, each character represented as a shared character embedding combined with a language-embedding vector, so that the same written letter can mean different sounds in different languages. The decoder generates the Latin proto-word character by character, using dot-product attention to pick out the most relevant cognate characters at each step, and an MLP with 200 hidden units produces the next character. The evaluation machinery is edit distance between the predicted and gold Latin form, reported as exact-match rate, average edit distance, and normalized edit distance.

What would settle it

Randomly sample test entries from the parts of the dataset that were not manually verified, have expert linguists check the cognate sets against etymological dictionaries and correct the IPA transcriptions by hand, then recompute the reported edit distances; if the corrections move the average orthographic edit distance from 0.65 by more than a small amount or push exact reconstruction below the earlier baseline, the central claim fails.

Watch

Extended reading notes

Core claim

The central claim is that a supervised character-level encoder-decoder with attention can perform proto-word reconstruction well enough to beat earlier computational approaches, and that the model's internal representations and errors correspond to real historical phonology. Trained on sets of cognates tagged with their languages, the model reads the daughter forms and emits the Latin ancestor one character at a time. On the orthographic test set it reaches 64.1% exact reconstruction, 84.0% within one edit operation, with average edit distance 0.65 and normalized distance 0.064; the phonetic variant reaches 50.0% exact and normalized distance 0.100, with the gap largely explained by Latin tense-lax vowel contrasts that daughter languages neutralize. The paper further claims that roughly 80% of orthographic and 75% of phonetic errors fall into linguistically named categories, that the network correctly predicts 22 of 33 documented phonological-change rules on a synthetic test, and that its learned phoneme embeddings cluster into a phonologically meaningful hierarchy (vowels vs. consonants, voiced vs. voiceless pairs, allophones) without explicit supervision.

Load-bearing premise

The results stand on the dataset's correctness: the Wiktionary-derived cognate sets were manually checked only in part and the automatic IPA transcriptions were verified on only 200 words, so systematic errors in the unverified entries would distort every reported reconstruction score and every conclusion about learned sound changes.

Editorial extensions

If this is right

  • For well-attested language families, large-scale reconstruction of proto-lexicons becomes a semi-automatic process: linguists supply or verify cognate sets, and the model proposes ancestor forms that can be checked.
  • The systematic error categories make the model a diagnostic instrument: where reconstruction fails, the failure identifies sound changes that daughter languages have made opaque, indicating where extra evidence such as inscriptions or other branches is needed.
  • The phonologically structured embeddings suggest that training on reconstruction tasks is a way to learn phoneme taxonomies without labeled data, which could transfer to under-resourced languages.
  • The public 8,799-entry dataset gives the field a common benchmark, so future reconstruction methods can be compared on the same cognate sets and metrics.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the pattern generalizes, the same architecture could be tested on other attested proto-languages, such as Germanic or Slavic families, to see whether the 64% exact-reconstruction rate is typical or specific to the conservative orthography and close relatedness of Romance languages.
  • Because the model's success depends on having access to the full cognate set, a natural next step is to couple it with automatic cognate detection and measure how reconstruction accuracy degrades as cognate sets become noisier or incomplete.
  • The attention analysis suggests a concrete, testable extension: if a daughter language is most attended at a position, removing its cognate should hurt reconstruction at that position more than removing a low-attention language's cognate; this prediction could be checked with ablation experiments.
  • If model errors reliably flag opaque changes, the same approach could be used to triage large dictionaries for human etymological review, making the comparative method faster without replacing the linguist.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper addresses automated proto-word reconstruction: given cognate forms in five Romance daughter languages, a character-level neural encoder-decoder predicts the Latin proto-word. The authors contribute a new dataset of 8,799 comparative entries, combining a cleaned version of the Ciobanu and Dinu (2014b) dataset with Wiktionary-derived additions, in both orthographic and eSpeak-transcribed IPA forms. On a held-out test set, the orthographic model achieves an average edit distance of 0.65, normalized edit distance of 0.064, and 64.1% exact reconstructions, which the authors compare favorably to the 1.07 average edit distance reported by Ciobanu and Dinu (2018). The paper also provides an error taxonomy, a synthetic test of 33 documented phonological-change rules, a hierarchical clustering of learned phoneme embeddings, and an attention analysis across languages.

Significance. If the results hold, this is a valuable contribution to computational historical linguistics. The new dataset is substantially larger than existing resources and the cleaning protocol demonstrably improves data quality (43 errors in a 300-entry sample of the original data versus 4 in the cleaned data). The synthetic rule evaluation, in which the model correctly handles 22 of 33 documented sound changes, and the learned-embedding analysis provide evidence that the model internalizes phonologically meaningful regularities rather than merely memorizing surface correspondences. The paper is also refreshingly explicit about the limits of the supervised setting relative to the historical linguist's task. However, the strength of these conclusions is tempered by evaluation-protocol issues: the headline comparison to prior work is cross-dataset, the test set is not fully reproducible from the public release, and the ground-truth labels rest on relatively small manual verification samples.

major comments (3)
  1. [Section 6, Table 1] The abstract and Section 6 claim that neural sequence models outperform conventional methods 'applied to this task so far,' but the only quantitative basis is a cross-dataset comparison to Ciobanu and Dinu (2018), who evaluated on a different, unreleased test set. No baseline is run on the new 1,055-entry test set, so the reader cannot tell whether the 0.65 edit distance reflects model quality or dataset ease. Footnote 7 mitigates the concern by reporting a smaller model on the original dataset (0.881 vs. 1.07), but that is still a comparison across implementations and hyperparameters. Add same-data baselines to Table 1, such as the CRF of Ciobanu and Dinu (2018) re-run on the new dataset or a simple alignment/majority baseline, to support the superiority claim on a common benchmark.
  2. [Section 4 and footnote 6] The test split of 1,055 entries is not fully reproducible: the release excludes entries that appeared in the original Ciobanu and Dinu (2014b) dataset, and the random split is not seeded or otherwise described. Moreover, all reported metrics come from a single training run, with no variance estimate. Because the headline numbers (0.65 average edit distance, 64.1% exact match) are the paper's central quantitative claim, the authors should report results over multiple random seeds (mean and standard deviation), state the seed used for the split, and either release the full test set (including cleaned original entries if legally possible) or release the model's test-set predictions so the exact numbers can be independently verified.
  3. [Appendix A.1] The quality of the ground-truth labels is established by manual checks of only 300 entries for error counting and 200 IPA transcriptions out of 8,799 entries and over 41,000 distinct words. The paper's own footnote 7 shows that preprocessing decisions of the kind in question change average edit distance by more than 0.2 units (0.881 vs. 0.612 on the original dataset), so label noise is not a negligible factor. A structured audit of a larger sample, stratified by source (Ciobanu-Dinu vs. Wiktionary) and by language, is needed to rule out systematic errors in the unverified majority, particularly for the Wiktionary additions and the eSpeak vowel-length and quality transcriptions.
minor comments (6)
  1. [Table 1] The table header contains typos: 'Ortographic' should be 'Orthographic' and 'datsaets' should be 'datasets'.
  2. [Section 6.1] In the paragraph on 'Other vowel changes,' 'This lef to reconstruction errors' should read 'This led to reconstruction errors'; also, 'A separated case is that of Greek words' should be 'A separate case.'
  3. [Section 6] In the discussion of vowel-length datasets, 'lenghts' should be 'lengths' (two occurrences).
  4. [Section 3] The running example uses 'lactem' as the Latin word for 'milk'; classical Latin is 'lac' (genitive 'lactis'). If the dataset normalizes to a later or analogical accusative form, this should be stated explicitly for readers unfamiliar with the normalization.
  5. [Section 5.1] Training details such as number of epochs, learning rate, optimizer, batch size, and early-stopping criterion are not reported; these details are needed for reproducibility.
  6. [Section 6.5 / Figure 4] The figure caption and text should define precisely how the counts are normalized ('with respect to time step, letter frequency and language frequency in the corpus') and state whether the reported values are percentages or ratio scores.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the central reconstruction result is a supervised held-out evaluation, and the supporting analyses use externally sourced phonological rules and unsupervised representations.

full rationale

The paper's central claim is empirical: a character-level GRU encoder-decoder, trained on (cognate-set, Latin proto-form) pairs, predicts held-out Latin forms with an average edit distance of 0.65. The test split of 1,055 entries is disjoint from the training split, and the target labels are gold Latin forms from Ciobanu and Dinu (2014b) plus manually checked Wiktionary entries, not outputs of the model. The comparison to Ciobanu and Dinu (2018) is explicitly hedged in footnote 7, which reports re-runs on the original and cleaned datasets, so the improvement claim is not an imported conclusion. The error taxonomy in Table 2 is a post hoc analysis of model outputs rather than an input to training. The synthetic-rule evaluation in Section 6.3 uses 33 phonological rules drawn from external historical-linguistics references (Boyd-Bowman 1980; Alkire and Rosen 2010) and evaluates isolated syllables distinct from the training distribution. The phoneme-embedding analysis is an unsupervised clustering of the model's own learned representations, and the attention analysis inspects the model's behavior without imposing the conclusion. The main caveats—limited manual verification of Wiktionary and eSpeak labels, the non-public release of the original Ciobanu-Dinu entries, and the absence of seed-variance reporting—are data-quality and reproducibility concerns, not circularity. No equation in the paper reduces to its own input, and no load-bearing result is justified solely by a self-citation.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The paper is an empirical ML contribution. It introduces no new theoretical entities or fitted physical constants. The central claims rest on the standard supervised-learning setup plus the domain-specific assumption that sound change is regular, and on the quality of a dataset that is only partially manually verified.

free parameters (4)
  • Encoder/decoder hidden size = 150
    GRU hidden size in the character-level encoder-decoder; chosen by hand without reported sensitivity analysis. Differences in this choice can change exact accuracies.
  • Character embedding size = 100
    Embedding dimension for characters and languages; a standard hyperparameter choice without reported tuning.
  • MLP hidden units = 200
    Hidden layer size in the attention output MLP; chosen by hand.
  • Train/eval/test split = 80/8/12%
    Random split proportions stated in Section 4; no random seed is reported, so the exact test set is not defined by the paper.
assumptions (4)
  • domain assumption Regularity of sound change
    The task is learnable only if historical sound changes are systematic and predictable from daughter languages, as stated in Section 1: "historical sound changes within a language family are not random."
  • domain assumption Accuracy of eSpeak IPA transcriptions
    The phonetic dataset is generated by automatic eSpeak transcription (Sections 4 and A.1); a manual check of 200 words found them correct after global fixes, but the remaining thousands of transcriptions are unverified.
  • domain assumption Correctness of cognate annotations and Latin lemmas
    The dataset combines Ciobanu and Dinu (2014b) entries with Wiktionary scraping. Only 300 entries were manually sampled to estimate error rates (Section A.1); the evaluation assumes the rest are correct.
  • standard math Representativeness of the random split
    The 80/8/12 random split is assumed to give an unbiased test distribution; the paper does not report the split seed or check for lemma overlap between train and test.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Ab Antiquo: Neural Proto-language Reconstruction." pith.science (2026). https://pith.science/paper/DSAUU4GS

@misc{pith2026190802477,
  author       = {Pith},
  title        = {Pith review of: Ab Antiquo: Neural Proto-language Reconstruction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DSAUU4GS}},
  note         = {Machine review of arXiv:1908.02477}
}
read the original abstract

Historical linguists have identified regularities in the process of historic sound change. The comparative method utilizes those regularities to reconstruct proto-words based on observed forms in daughter languages. Can this process be efficiently automated? We address the task of proto-word reconstruction, in which the model is exposed to cognates in contemporary daughter languages, and has to predict the proto word in the ancestor language. We provide a novel dataset for this task, encompassing over 8,000 comparative entries, and show that neural sequence models outperform conventional methods applied to this task so far. Error analysis reveals variability in the ability of neural model to capture different phonological changes, correlating with the complexity of the changes. Analysis of learned embeddings reveals the models learn phonologically meaningful generalizations, corresponding to well-attested phonological shifts documented by historical linguistics.

Figures

Figures reproduced from arXiv: 1908.02477 by the authors.

Figure 1
Figure 1. the evolution of Latin word for “sky” is sev [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Phonological mistakes resulting from alter [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Hierarchical clustering of French phoneme [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: position in output vs. most attended lan [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 32 canonical work pages

  1. [1]

    URL: " 'urlintro :=

    ENTRY address author booktitle chapter edition editor howpublished institution journal key month note number organization pages publisher school series title type volume year eprint doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRINGS urlintro eprinturl eprintpr...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    , " * write output.state after.block = add.period write newline

    ENTRY address author booktitle chapter edition editor howpublished institution journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.all := #1 'mid.sentence := #2 '...

  4. [4]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  5. [5]

    Ti Alkire and Carol Rosen. 2010. Romance languages: A historical introduction. Cambridge University Press

    work page 2010

  6. [6]

    W Sidney Allen and William Sidney Allen. 1989. Vox latina. Cambridge University Press

    work page 1989

  7. [7]

    Raimo Anttila. 1989. Historical and comparative linguistics, volume 6. John Benjamins Publishing

    work page 1989

  8. [8]

    Dzmitry Bahdanau, Kyunghyun Cho, and Yoshua Bengio. 2015. http://arxiv.org/abs/1409.0473 Neural machine translation by jointly learning to align and translate . In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings

    arXiv 2015

Show all 41 references
  1. [9]

    Alexandre Bouchard-C \^o t \'e , Thomas L Griffiths, and Dan Klein. 2009. Improved reconstruction of protolanguage word forms. In Proceedings of human language technologies: The 2009 annual conference of the north american chapter of the association for computational linguisti...

    2009

  2. [10]

    Alexandre Bouchard-C \^o t \'e , David Hall, Thomas L Griffiths, and Dan Klein. 2013. Automated reconstruction of ancient languages using probabilistic models of sound change. Proceedings of the National Academy of Sciences, 110(11):4224--4229

    2013

  3. [11]

    Griffiths, and Dan Klein

    Alexandre Bouchard - C \^ o t \' e , Percy Liang, Thomas L. Griffiths, and Dan Klein. 2007. http://www.aclweb.org/anthology/D07-1093 A probabilistic approach to diachronic phonology . In EMNLP-CoNLL 2007, Proceedings of the 2007 Joint Conference on Empirical Methods in Natural...

    2007

  4. [12]

    Peter Boyd-Bowman. 1980. From Latin to Romance in sound charts. Georgetown University Press

    1980

  5. [13]

    Kyunghyun Cho, Bart van Merrienboer, Dzmitry Bahdanau, and Yoshua Bengio. 2014. http://aclweb.org/anthology/W/W14/W14-4012.pdf On the properties of neural machine translation: Encoder-decoder approaches . In Proceedings of SSST@EMNLP 2014, Eighth Workshop on Syntax, Semantics ...

    2014

  6. [14]

    Alina Maria Ciobanu and Liviu P Dinu. 2014 a . Automatic detection of cognates using orthographic alignment. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), volume 2, pages 99--105

    2014

  7. [15]

    Alina Maria Ciobanu and Liviu P Dinu. 2014 b . Building a dataset of multilingual cognates for the romanian lexicon. In Proceedings of the 9th International Conference on Language Resources and Evaluation, LREC, pages 1038--1043

    2014

  8. [16]

    Alina Maria Ciobanu and Liviu P Dinu. 2018. Ab initio: Automatic latin proto-word reconstruction. In Proceedings of the 27th International Conference on Computational Linguistics, pages 1604--1614

    2018

  9. [17]

    J Halvor Clegg and Willis C Fails. 2017. Manual de fon \'e tica y fonolog \' a espa \ n olas . Routledge

    2017

  10. [18]

    Josette Rey Debove and Alain Rey. 2000. Le Nouveau Petit Robert: Dictionnaire alphabetique et analogique de la langue francaise. Dictionnaires Le Robert

    2000

  11. [19]

    Friedrich Christian DIEZ and TC Donkin. 1864. An Etymological Dictionary of the Romance Languages; chiefly from the German of F. Diez. By TC Donkin. Williams and Norgate

  12. [20]

    Gerhard Ernst. 2003. Romanische Sprachgeschichte: Ein Internationales Handbuch Zur Geschichte Der Romanischen Sprachen Und Ihrer Erforschung/Manuel International Sur L'Histoire Et L'Etude Linguistique DES Langues Romanes. Mouton de Gruyter

    2003

  13. [21]

    F \'e lix Gaffiot and Pierre Flobert. 1934. Dictionnaire latin-fran c ais . Hachette Paris

    1934

  14. [22]

    Robert A Hall. 1944. Italian phonemes and orthography. Italica, 21(2):72--82

    1944

  15. [23]

    Martin Harris and Nigel Vincent. 2003. The romance languages. Routledge

    2003

  16. [24]

    G \"u nter Holtus, Michael Metzeltin, and Christian Schmitt. 1989. LRL, volume 3. M. Niemeyer

    1989

  17. [25]

    Diana Inkpen, Oana Frunza, and Grzegorz Kondrak. 2005. Automatic identification of cognates and false friends in french and english. In Proceedings of the International Conference Recent Advances in Natural Language Processing, volume 9, pages 251--257

    2005

  18. [26]

    Grzegorz Kondrak. 2001. Identifying cognates by phonetic and semantic similarity. In Proceedings of the second meeting of the North American Chapter of the Association for Computational Linguistics on Language technologies, pages 1--8. Association for Computational Linguistics

    2001

  19. [27]

    Adam Ledgeway and Martin Maiden. 2016. The Oxford guide to the Romance languages, volume 1. Oxford University Press

    2016

  20. [28]

    Lewis and Charles Short

    Charlton T. Lewis and Charles Short. 1879. A latin dictionary. Perseus Digital Library

  21. [29]

    McCarthy, and David Yarowsky

    Dylan Lewis, Winston Wu, Arya D. McCarthy, and David Yarowsky. 2020. https://doi.org/10.18653/v1/2020.coling-main.387 Neural transduction for multilingual lexical translation . In Proceedings of the 28th International Conference on Computational Linguistics, COLING 2020, Barce...

    2020 doi

  22. [30]

    Johann-Mattis List, Philippe Lopez, and Eric Bapteste. 2016. Using sequence similarity networks to identify partial cognates in multilingual wordlists. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), volume 2...

    2016

  23. [31]

    Mann and David Yarowsky

    Gideon S. Mann and David Yarowsky. 2001. http://aclweb.org/anthology/N/N01/N01-1020.pdf Multipath translation lexicon induction via bridge languages . In Language Technologies 2001: The Second Meeting of the North American Chapter of the Association for Computational Linguisti...

    2001

  24. [32]

    Maria Helena Mateus and Ernesto d'Andrade. 2000. The phonology of Portuguese. OUP Oxford

    2000

  25. [33]

    Robert McColl Millar. 2013. Trask's historical linguistics. Routledge

    2013

  26. [34]

    Andrea Mulloni and Viktor Pekar. 2006. Automatic detection of orthographics cues for cognate recognition. In LREC, pages 2387--2390

    2006

  27. [35]

    Yuta Nishimura, Katsuhito Sudoh, Graham Neubig, and Satoshi Nakamura. 2020. https://doi.org/10.1109/TASLP.2019.2959224 Multi-source neural machine translation with missing data . IEEE ACM Trans. Audio Speech Lang. Process. , 28:569--580

    2020

  28. [36]

    Fabian Pedregosa, Ga \"e l Varoquaux, Alexandre Gramfort, Vincent Michel, Bertrand Thirion, Olivier Grisel, Mathieu Blondel, Peter Prettenhofer, Ron Weiss, Vincent Dubourg, et al. 2011. Scikit-learn: Machine learning in python. Journal of machine learning research, 12(Oct):2825--2830

    2011

  29. [37]

    Taraka Rama, Johann - Mattis List, Johannes Wahle, and Gerhard J \" a ger. 2018. https://aclanthology.info/papers/N18-2063/n18-2063 Are automatic methods for cognate detection good enough for phylogenetic reconstruction in historical linguistics? In Proceedings of the 2018 Con...

    2018

  30. [38]

    Mika Sarlin. 2014. Romanian grammar. BoD-Books on Demand

    2014

  31. [39]

    Joe H Ward Jr. 1963. Hierarchical grouping to optimize an objective function. Journal of the American statistical association, 58(301):236--244

    1963

  32. [40]

    Winston Wu, Garrett Nicolai, and David Yarowsky. 2020. https://www.aclweb.org/anthology/2020.lrec-1.519/ Multilingual dictionary based construction of core vocabulary . In Proceedings of The 12th Language Resources and Evaluation Conference, LREC 2020, Marseille, France, May 1...

    2020

  33. [41]

    Winston Wu and David Yarowsky. 2018. http://www.lrec-conf.org/proceedings/lrec2018/summaries/934.html Creating large-scale multilingual cognate tables . In Proceedings of the Eleventh International Conference on Language Resources and Evaluation, LREC 2018, Miyazaki, Japan, Ma...

    2018

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.