Parsimonious Morpheme Segmentation with an Application to Enriching Word Embeddings

Ahmed El-Kishky; Aston Zhang; Frank Xu; Jiawei Han

REVIEW 3 major objections 5 minor 45 references

Parsimonious Morpheme Segmentation with an Application to Enriching Word Embeddings

T0 review · 3 major / 5 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 Unsupervised segmentation splits words into the fewest morphemes, improving rare-word embeddings.

desk verdict Solid segmentation core, oversold embedding claims, and an untested entropy-boundary premise that caps recall. read the letter →

arxiv 1908.07832 v2 pith:5FWOQWKA submitted 2019-08-18 cs.CL cs.LGstat.ML

Ahmed El-Kishky , Frank Xu , Aston Zhang , Jiawei Han This is my paper · ORCID

classification cs.CLcs.LGstat.ML

keywords morphemesegmentationunsupervisedwordembeddingssubwordinformationout-of-vocabularywordsentropyboundarydetectionparsimonycriterionmulti-granularity

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 claims that unsupervised morpheme segmentation can be driven by a parsimony principle: at every level of a hierarchy, segment each word into the fewest morphemes that cover it, using candidate boundaries found by information entropy. It presents MorphMine, which identifies prefix and suffix boundaries at local entropy maxima, extracts shared roots, segments via dynamic programming, then globally resegments to favor morphemes used elsewhere in the vocabulary. Evaluations on English, German, and Turkish report higher F1 than BPE, unigram LM, and Morfessor on MorphoChallenge, and enriched embeddings improve similarity, analogy, and language-model perplexity, especially for rare and out-of-vocabulary words. The significance is a label-free, language-general way to share morphological structure and to build embeddings for long-tail words.

What carries the argument

The central mechanism is an entropy-boundary detector coupled with a dynamic-programming solver for Disjoint Interval Covering: choose the smallest set of non-overlapping candidate morpheme intervals that maximizes character coverage. Prefix and suffix candidates come from local maxima in the entropy of the next-character distribution, and root candidates arise by stripping these affixes. The DP selects parsimonious segmentations, maximum likelihood breaks ties among them, and a global resegmentation pass recomputes morpheme counts and re-segments every word, aligning choices toward morphemes that are shared across the vocabulary.

What would settle it

Run MorphMine on a language with non-concatenative morphology, such as Arabic root-and-pattern morphology, or on a synthetic vocabulary where boundaries are known but transition entropy is deliberately flat; if gold-segmentation F1 drops to the level of the baselines or below, the entropy-boundary premise is the load-bearing failure point.

Watch

Extended reading notes

Core claim

MorphMine's thesis is that parsimony — choosing the fewest morphemes that maximally cover each word — is the right inductive bias for unsupervised morpheme segmentation, provided candidate boundaries come from transition-entropy peaks. Segmenting hierarchically (words into morphemes, then morphemes into sub-morphemes) and resegmenting globally with refined counts yields human-verifiable morphemes that outperform existing unsupervised segmenters. Using the resulting multi-granular morphemes in a FastText-style additive embedding model improves rare-word similarity, syntactic analogy performance, and held-out language-model perplexity on English, German, and Turkish.

Load-bearing premise

A local peak in character-transition entropy marks a true morpheme boundary, so if a language's morphology does not create such entropy peaks, the candidate pool is wrong and the parsimonious dynamic program cannot recover real morphemes.

Editorial extensions

If this is right

MorphMine provides a parameter-free, language-general preprocessing step: no morpheme count, no linguistic rules, and linear-time scaling with vocabulary size.
Multi-granular morphemes improve embeddings for rare and out-of-vocabulary words because longer shared morphemes link rare forms to frequent ones.
Morpheme-enriched embeddings also lower cross-entropy perplexity of a downstream LSTM language model on English, German, and Turkish.
The parsimony criterion yields longer morphemes than BPE, unigram LM, or Morfessor, preserving semantic ties among derived forms such as truncate, truncated, and truncating.

Reading between the lines

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

The entropy-boundary premise could be tested against other boundary signals, such as minimum-description-length breaks or predictability drops; if a stronger signal exists, the same parsimonious covering DP would likely improve further.
Because MorphMine only needs a vocabulary list, not a corpus, it could be applied to specialized quasi-concatenative vocabularies such as biomedical terms or chemical names, where word forms are long and rare.
The global resegmentation pass is an expectation-maximization-like procedure over morpheme counts; iterating to convergence rather than one pass could be measured for stability or over-segmentation.
For non-concatenative morphology, the same parsimony objective would require a different candidate generator, since interleaved root-and-pattern morphemes do not produce clean transition-entropy peaks.

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.

Desk Editor's Note

read the letter

Short version: MorphMine is a genuinely useful segmentation system with a clean algorithmic core. The segmentation results are the strong part; the embedding claims are oversold; and the entropy-boundary premise is an untested ceiling on recall.

What's new: the paper formalizes morpheme segmentation as disjoint interval covering with a parsimony objective and solves it with a linear-time DP, adds a global resegmentation pass, and produces multi-granular segmentations. The F1 gains over Morfessor, BPE, and unigram LM on MorphoChallenge are concrete: English 0.760 vs 0.699, German 0.662 vs 0.615, Turkish 0.542 vs 0.483. That is a real improvement, and the case study shows MorphMine captures larger roots like 'vandal' and 'truncat' that baselines miss.

The stress-test note about the entropy-boundary premise is right. Section III-A defines candidate prefixes and suffixes only at local entropy maxima, so the DP can never propose a boundary that is not an entropy peak. The paper never checks how many gold boundaries satisfy that condition; the one example in Figure 2 is anecdotal. The recall ceiling likely explains part of the drop from English to Turkish, where vowel harmony makes suffix boundaries less surprising. This does not sink the method—it still beats Morfessor on all three languages—but it should be measured or stated as a limitation.

The embedding enrichment is the weakest section. Table III shows MorphMine does not consistently beat SkipGram on semantic analogies (English 67 vs 68), yet the abstract says 'consistently improves.' There are no error bars or significance tests, and the BPE/ULM baselines were tuned on the test set. The perplexity gains are small (154 vs 155 for Morfessor on English) and may not be meaningful. No code is released, so the threshold parameters (min root support, min root length) cannot be checked.

Who this is for: anyone working on morphology or subword embeddings will get useful ideas, especially the DP formulation and the global resegmentation trick. The paper deserves serious peer review, but the authors need to either add code/data and proper significance testing or tone down the embedding claims.

Referee Report

3 major / 5 minor

Summary. The paper proposes MorphMine, an unsupervised morpheme segmentation method that constructs a candidate morpheme vocabulary from local maxima in prefix/suffix character-transition entropy and from residual root strings, then segments each word with a parsimonious dynamic-programming algorithm (Disjoint Interval Covering) that selects the fewest candidate morphemes covering the word, with maximum-likelihood tie-breaking and a global resegmentation step. The extracted morphemes are then used to enrich FastText-style word embeddings. Experiments on English, German, and Turkish report F1 improvements over Morfessor, byte-pair encoding, and unigram language-model segmentations on MorphoChallenge data, and claim improved word-similarity, analogy, and language-model perplexity results.

Significance. If the segmentation results are robust, the method would be a meaningful advance: it is fully unsupervised, language-agnostic, and achieves the best reported F1 on three languages (e.g., English 0.760 vs. 0.699 for Morfessor) without external linguistic resources. The hierarchical, multi-granular segmentation and its integration into embeddings though parameter sharing is a plausible mechanism for improving representations of rare and out-of-vocabulary words. The algorithmic description, including the recurrence in Eq. (1) and Algorithms 1 and 2, is detailed enough to permit replication, although no code or data release is mentioned.

major comments (3)

[Section III-A; Fig. 2; Table I] The entire candidate morpheme vocabulary is generated from the premise that a true morpheme boundary appears as a local maximum in prefix/suffix transition entropy; the DP in Eq. (1) and Algorithm 2 can only assemble intervals from this pool, so the method's recall is capped by the fraction of gold boundaries that are entropy maxima. Figure 2 is a single anecdote, and no systematic check is reported. The declining F1 from English (0.760) to German (0.662) to Turkish (0.542) is consistent with this premise weakening for agglutinative morphology, where vowel harmony can make suffix-initial characters predictable and suppress entropy peaks. Please add an oracle-recall experiment on the MorphoChallenge data: of the gold segment boundaries, what fraction are local maxima of the proposed entropy, and how much of the final recall is lost by the DP relative to that oracle?
[Section V-B; Section V-C; Abstract] The abstract's claim that enriching embeddings with MorphMine morphemes "consistently improves embedding quality" is contradicted by Table III: on English semantic analogies MorphMine scores 67 versus 68 for SkipGram, and on German semantic analogies 61 versus 63; the consistent gains are confined to syntactic analogies (78 vs. 75 English, 53 vs. 46 Turkish) and to some similarity sets. Moreover, no error bars, significance tests, or multiple-seed runs are reported, and several differences in Table II are within one point. Please qualify the claim to the tasks where gains are consistent and add statistical significance or confidence intervals, or at minimum state the number of runs averaged.
[Section V-A; Table I] The BPE and unigram-LM baselines are reported from a parameter sweep over vocabulary size, taking the "highest performing run," while MorphMine's own thresholds (minimum root support 2, minimum root length 4, and minimum morpheme frequency f(m)>1) are fixed. This creates a selection bias and makes the headline F1 differences hard to interpret without variance information. Please report the mean and standard deviation of the baselines across the sweep, or a sensitivity curve, and clarify whether the MorphoChallenge F1 is computed only on the gold-annotated words or on the full vocabulary.

minor comments (5)

[Section III, first paragraph] The text states that the method is "non-parametric" but later imposes minimum root support of 2, minimum root length of 4 (Section III-A2), and a minimum morpheme frequency of f(m)>1 for tie-breaking (Section III-B1); please remove the "non-parametric" claim or explicitly list these as tunable thresholds.
[Abstract; Section V] There are several typographical errors: "variety of of" in the abstract, "emebeddings" in Section V, "backstracking" in Section III-B, "mostly likely" in Example 3, "completness" in Example 3, "mutliple" in Section V-D, and "were" for "where" in Section V-B; a careful proofread is needed.
[Section V-F; Fig. 4] The scalability regressions report R^2 values of 0.989 and 0.991 but do not state the number of data points, the vocabulary sizes sampled, or whether each point is a single run; please describe the runtime measurement setup.
[Section V-B; Table II] Table II labels columns "RW-Frequent" and "RW-OOV," but the dataset description in Section V only mentions the Stanford Rare Words (RW) set; please clarify the distinction between the frequent and OOV splits and how OOV words are evaluated.
[Section III-B; Eq. (1)] The typeset recurrence in Eq. (1) has garbled max/min braces and unexplained 0/1 subscripts; please present the DP recurrence in a cleaner and more standard form, with all symbols defined.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: MorphMine is an unsupervised segmentation algorithm evaluated against external gold-standard MorphoChallenge data and independent downstream benchmarks.

full rationale

MorphMine's derivation is self-contained. The candidate morpheme vocabulary is generated by an entropy local-maximum heuristic over character-transition distributions (Section III-A), and the DP in Section III-B optimizes coverage and parsimony over those candidates. This is a stated modeling assumption, not a definition of the gold output; the MorphoChallenge ground-truth segmentations are external and never used in fitting. The global resegmentation (Section III-D) does recompute morpheme counts from the algorithm's own first-pass output and uses them as likelihood tie-breakers, which is a self-referential bootstrap, but it is not a derivation of the paper's evaluation claims: the reported F1, similarity, analogy, and perplexity numbers are measured against external or held-out data. The only self-citation ([11]) is an acknowledgment that MorphMine continues preliminary work; the entropy-boundary equations and DP recurrence are stated in the paper itself, so no load-bearing claim reduces to the self-citation. The entropy-boundary premise is a testable assumption that may limit recall for some languages, but that is a correctness or robustness concern, not circularity.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

MorphMine rests on a small set of hand-set thresholds and linguistic assumptions, and none of the thresholds has a sensitivity analysis. No new physical or conceptual entities are postulated; the multi-granular hierarchy is a representation, not an invented entity.

free parameters (3)

minimum root support = 2
Section III-A.2: root candidates require a minimum support of two within the vocabulary. This hand-set threshold controls the candidate pool and has no reported sensitivity analysis.
minimum root length = 4
Section III-A.2: root candidates require a minimum length of four characters. This hand-set threshold directly affects which roots are available for segmentation, and no sensitivity analysis is reported.
minimum morpheme frequency in tie-break = >1
Section III-B.1: the selected most-likely partition cannot contain morphemes that occur only once in the vocabulary. This frequency floor prunes rare morphemes and shapes the final segmentation.

assumptions (6)

domain assumption Morpheme compositionality: the vocabulary is generated by concatenating morphemes drawn from a smaller morpheme vocabulary.
Invoked at the start of Section III-A and in Algorithm 2. If words are not compositional, the candidate-generation and segmentation framing fails.
domain assumption A local maximum in character-transition entropy marks a morpheme boundary.
Section III-A, entropy scoring and Figure 2. This empirical linguistic heuristic generates all prefix and suffix candidates and is not derived from first principles.
domain assumption Parsimony: the best segmentation is the one with the fewest morphemes that maximally cover the word.
Section III-B and Equation 1. Occam's razor is posited as the segmentation criterion, not proven to recover linguistic morphemes.
domain assumption Independence of morphemes in maximum-likelihood tie-breaking.
Section III-B.1: the likelihood is computed as a product of morpheme frequencies, which assumes morphemes are independent within a word.
domain assumption Semantically meaningful morphemes improve word embeddings when summed into word vectors.
Section III-E: adapting FastText to sum MorphMine morphemes assumes this relationship, and it is only supported empirically.
standard math The recurrence in Equation 1 computes an optimal disjoint interval covering.
The paper states the recurrence without a formal proof of optimality. The memoized DP is standard, but no correctness proof is provided.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Parsimonious Morpheme Segmentation with an Application to Enriching Word Embeddings." pith.science (2026). https://pith.science/paper/5FWOQWKA

@misc{pith2026190807832,
  author       = {Pith},
  title        = {Pith review of: Parsimonious Morpheme Segmentation with an Application to Enriching Word Embeddings},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5FWOQWKA}},
  note         = {Machine review of arXiv:1908.07832}
}

read the original abstract

Traditionally, many text-mining tasks treat individual word-tokens as the finest meaningful semantic granularity. However, in many languages and specialized corpora, words are composed by concatenating semantically meaningful subword structures. Word-level analysis cannot leverage the semantic information present in such subword structures. With regard to word embedding techniques, this leads to not only poor embeddings for infrequent words in long-tailed text corpora but also weak capabilities for handling out-of-vocabulary words. In this paper we propose MorphMine for unsupervised morpheme segmentation. MorphMine applies a parsimony criterion to hierarchically segment words into the fewest number of morphemes at each level of the hierarchy. This leads to longer shared morphemes at each level of segmentation. Experiments show that MorphMine segments words in a variety of languages into human-verified morphemes. Additionally, we experimentally demonstrate that utilizing MorphMine morphemes to enrich word embeddings consistently improves embedding quality on a variety of of embedding evaluations and a downstream language modeling task.

Figures

Figures reproduced from arXiv: 1908.07832 by the authors.

**Figure 1.** Hierarchical segmentation of words. many methods have been developed to learn these word representations from large, unlabeled text corpora [5], [29], [30]. Despite many advances, unsupervised learning of distributed representations can struggle in learning adequate vectors for infrequent words. This problem is ubiquitous because most text corpora demonstrate long-tail distributions in relation to word frequency, wi… view at source ↗

**Figure 2.** Prefix and suffix transition entropy. trial, n = 1, we simplify it into entropy of the categorical distribution. For the prefix mi : H(mi) = − X C j=1 P(mi ⊕ cj |mi) × log2P(mi ⊕ cj |mi), where ⊕ denotes the binary string concatenation of two strings and the transitional prefix probability is estimated as: P(mi ⊕ cj |mi) = f(mi ⊕ cj ) f(mi) and f(mi) denotes the frequency of a prefix mi in the input vocabulary list.… view at source ↗

**Figure 3.** Segmentation of the word “spatiotemporal” using disjoint interval covering. 2) Root Word Generation: Utilizing entropy-scoring, it is possible to detect morpheme structures that occur at the beginning or end of a word. However, many words often contain morpheme structure between prefixes and suffixes. For each prefix and suffix candidate identified in a word, it is possible to generate many candidate root words by s… view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Decomposition of morpheme segmentation algorithm into unsupervised morpheme mining then vocabulary segmentation. VI. CONCLUSIONS In this study, we propose a pattern-mining method of segmenting vocabulary into smaller morphemes and demonstrate experimentally on three …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 42 canonical work pages

[1]

Factored neural language models

Andrei Alexandrescu and Katrin Kirchhoff. Factored neural language models. In NAACL-HLT, pages 1–4. ACL, 2006

work page 2006
[2]

Semr-11: A multi-lingual gold-standard for semantic similarity and relatedness for eleven languages

Siamak Barzegar, Brian Davis, Manel Zarrouk, Siegfried Handschuh, and André Freitas. Semr-11: A multi-lingual gold-standard for semantic similarity and relatedness for eleven languages. In LREC, 2018

work page 2018
[3]

Enriching word vectors with subword information

Piotr Bojanowski, Edouard Grave, Armand Joulin, and Tomas Mikolov. Enriching word vectors with subword information. TACL, 2017

work page 2017
[4]

Alternative structures for character-level rnns

Piotr Bojanowski, Armand Joulin, and Tomas Mikolov. Alternative structures for character-level rnns. ICLR, 2015

work page 2015
[5]

A uniﬁed architecture for natural language processing: Deep neural networks with multitask learning

Ronan Collobert and Jason Weston. A uniﬁed architecture for natural language processing: Deep neural networks with multitask learning. In ICML, pages 160–167. ACM, 2008

work page 2008
[6]

Morphological word-embeddings

Ryan Cotterell and Hinrich Schütze. Morphological word-embeddings. In NAACL-HLT, pages 1287–1292, 2015

work page 2015
[7]

Unsupervised discovery of mor- phemes

Mathias Creutz and Krista Lagus. Unsupervised discovery of mor- phemes. In Proceedings of the ACL-02 workshop on Morphological and phonological learning-Volume 6 , pages 21–30. ACL, 2002

work page 2002
[8]

Knet: A general framework for learning word embedding using morphological knowledge

Qing Cui, Bin Gao, Jiang Bian, Siyu Qiu, Hanjun Dai, and Tie-Yan Liu. Knet: A general framework for learning word embedding using morphological knowledge. TOIS, 34(1):4, 2015

work page 2015

Show all 45 references

[9]

Morphemes as necessary concept for structures discovery from untagged corpora

Hervé Déjean. Morphemes as necessary concept for structures discovery from untagged corpora. In NeMLaP/CoNLL, pages 295–298. ACL, 1998

1998
[10]

Untangling herdan’s law and heaps’ law: Mathematical and informetric arguments

Leo Egghe. Untangling herdan’s law and heaps’ law: Mathematical and informetric arguments. Journal of the American Society for Information Science and Technology, 58(5):702–709, 2007

2007
[11]

Entropy-based subword mining with an application to word embeddings

Ahmed El-Kishky, Frank Xu, Aston Zhang, Stephen Macke, and Jiawei Han. Entropy-based subword mining with an application to word embeddings. In SCLeM, pages 12–21, 2018

2018
[12]

Finding structure in time

Jeffrey L Elman. Finding structure in time. Cognitive science , 14(2):179–211, 1990

1990
[13]

Anlamver: Semantic model evaluation dataset for turkish-word similarity and relatedness

Gökhan Ercan and Olcay Taner Yıldız. Anlamver: Semantic model evaluation dataset for turkish-word similarity and relatedness. In COLING, pages 3819–3836, 2018

2018
[14]

Problems with evaluation of word embeddings using word similarity tasks

Manaal Faruqui, Yulia Tsvetkov, Pushpendre Rastogi, and Chris Dyer. Problems with evaluation of word embeddings using word similarity tasks. In RepEval, pages 30–35, 2016

2016
[15]

Placing search in context: The concept revisited

Lev Finkelstein, Evgeniy Gabrilovich, Yossi Matias, Ehud Rivlin, Zach Solan, Gadi Wolfman, and Eytan Ruppin. Placing search in context: The concept revisited. ACM Transactions on information systems, 20(1):116– 131, 2002

2002
[16]

Scientiﬁc method in practice

Hugh G Gauch. Scientiﬁc method in practice . Cambridge University Press, 2003

2003
[17]

Word segmentation by letter successor varieties

Margaret A Hafer and Stephen F Weiss. Word segmentation by letter successor varieties. Information storage and retrieval , 10(11-12):371– 385, 1974

1974
[18]

Unsupervised learning of mor- phology

Harald Hammarström and Lars Borin. Unsupervised learning of mor- phology. Computational Linguistics, 37(2):309–350, 2011

2011
[19]

From phoneme to morpheme

Zellig S Harris. From phoneme to morpheme. In Papers in Structural and Transformational Linguistics , pages 32–67. Springer, 1970

1970
[20]

Simlex-999: Evaluating semantic models with (genuine) similarity estimation

Felix Hill, Roi Reichart, and Anna Korhonen. Simlex-999: Evaluating semantic models with (genuine) similarity estimation. Computational Linguistics, 41(4):665–695, 2015

2015
[21]

Bidirectional lstm-crf models for sequence tagging

Zhiheng Huang, Wei Xu, and Kai Yu. Bidirectional lstm-crf models for sequence tagging. arXiv preprint arXiv:1508.01991 , 2015

2015 arXiv
[22]

Bag of tricks for efﬁcient text classiﬁcation

Armand Joulin, Edouard Grave, Piotr Bojanowski, and Tomas Mikolov. Bag of tricks for efﬁcient text classiﬁcation. In EACL, pages 427–431, 2017

2017
[23]

Character-aware neural language models

Yoon Kim, Yacine Jernite, David Sontag, and Alexander M Rush. Character-aware neural language models. In AAAI, pages 2741–2749, 2016

2016
[24]

semantic

Maximilian Köper, Christian Scheible, and Sabine Schulte im Walde. Multilingual reliability and" semantic" structure of continuous word spaces. In IWCS, pages 40–45, 2015

2015
[25]

Mathematical linguistics

András Kornai. Mathematical linguistics. Springer Science & Business Media, 2007

2007
[26]

Subword regularization: Improving neural network trans- lation models with multiple subword candidates

Taku Kudo. Subword regularization: Improving neural network trans- lation models with multiple subword candidates. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 66–75, 2018

2018
[27]

Morpho challenge competition 2005–2010: evaluations and results

Mikko Kurimo, Sami Virpioja, Ville Turunen, and Krista Lagus. Morpho challenge competition 2005–2010: evaluations and results. In ACL Special Interest Group on Computational Morphology and Phonology , pages 87–95. ACL, 2010

2005
[28]

Better word representations with recursive neural networks for morphology

Thang Luong, Richard Socher, and Christopher Manning. Better word representations with recursive neural networks for morphology. In CoNLL, pages 104–113, 2013

2013
[29]

Efﬁcient estimation of word representations in vector space

Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. Efﬁcient estimation of word representations in vector space. arXiv preprint arXiv:1301.3781, 2013

2013 arXiv
[30]

Distributed representations of words and phrases and their compositionality

Tomas Mikolov, Ilya Sutskever, Kai Chen, Greg S Corrado, and Jeff Dean. Distributed representations of words and phrases and their compositionality. In NIPS, pages 3111–3119, 2013

2013
[31]

Linguistic regular- ities in continuous space word representations

Tomas Mikolov, Wen-tau Yih, and Geoffrey Zweig. Linguistic regular- ities in continuous space word representations. In NAACL-HLT, pages 746–751, 2013

2013
[32]

Unsupervised learning of morphology without morphemes

Sylvain Neuvel and Sean A Fulop. Unsupervised learning of morphology without morphemes. In Workshop on Morphological and phonological learning-Volume 6, pages 31–40. ACL, 2002

2002
[33]

Co-learning of word representations and morpheme representations

Siyu Qiu, Qing Cui, Jiang Bian, Bin Gao, and Tie-Yan Liu. Co-learning of word representations and morpheme representations. In COLING, pages 141–150, 2014

2014
[34]

Learn- ing representations by back-propagating errors

David E Rumelhart, Geoffrey E Hinton, Ronald J Williams, et al. Learn- ing representations by back-propagating errors. Cognitive modeling , 5(3):1, 1988

1988
[35]

Word segmentation: The role of distributional cues

Jenny R Saffran, Elissa L Newport, and Richard N Aslin. Word segmentation: The role of distributional cues. Journal of memory and language, 35(4):606–621, 1996

1996
[36]

Classiﬁcation of turkish semantic relation pairs using different sources

Gürkan Sahin. Classiﬁcation of turkish semantic relation pairs using different sources. International Journal of Computer Engineering and Information Technology, 8(10):196, 2016

2016
[37]

Morphology-based and sub-word language modeling for turkish speech recognition

Ha¸ sim Sak, Murat Saraclar, and Tunga Güngör. Morphology-based and sub-word language modeling for turkish speech recognition. In ICASSP, pages 5402–5405. IEEE, 2010

2010
[38]

Knowledge-free induction of inﬂectional morphologies

Patrick Schone and Daniel Jurafsky. Knowledge-free induction of inﬂectional morphologies. In NAACL, pages 1–9. ACL, 2001

2001
[39]

Japanese and korean voice search

Mike Schuster and Kaisuke Nakajima. Japanese and korean voice search. In ICASSP, pages 5149–5152. IEEE, 2012

2012
[40]

Neural machine translation of rare words with subword units

Rico Sennrich, Barry Haddow, and Alexandra Birch. Neural machine translation of rare words with subword units. In ACL, pages 1715–1725, 2016

2016
[41]

A mathematical theory of communication

Claude E Shannon. A mathematical theory of communication. ACM SIGMOBILE Mobile Computing and Communications Review , 5(1):3– 55, 2001

2001
[42]

Generating text with recurrent neural networks

Ilya Sutskever, James Martens, and Geoffrey E Hinton. Generating text with recurrent neural networks. In ICML, pages 1017–1024, 2011

2011
[43]

Learning sentiment-speciﬁc word embedding for twitter sentiment classiﬁcation

Duyu Tang, Furu Wei, Nan Yang, Ming Zhou, Ting Liu, and Bing Qin. Learning sentiment-speciﬁc word embedding for twitter sentiment classiﬁcation. In ACL, pages 1555–1565, 2014

2014
[44]

Google’s neural machine translation system: Bridging the gap between human and machine translation

Yonghui Wu, Mike Schuster, Zhifeng Chen, Quoc V Le, Mohammad Norouzi, Wolfgang Macherey, Maxim Krikun, Yuan Cao, Qin Gao, Klaus Macherey, et al. Google’s neural machine translation system: Bridging the gap between human and machine translation. arXiv preprint arXiv:1609.08144, 2016

2016 arXiv
[45]

Bilingual word embeddings for phrase-based machine translation

Will Y Zou, Richard Socher, Daniel Cer, and Christopher D Manning. Bilingual word embeddings for phrase-based machine translation. In EMNLP, pages 1393–1398, 2013

2013

Pith tools

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

[1] [1]

Factored neural language models

Andrei Alexandrescu and Katrin Kirchhoff. Factored neural language models. In NAACL-HLT, pages 1–4. ACL, 2006

work page 2006

[2] [2]

Semr-11: A multi-lingual gold-standard for semantic similarity and relatedness for eleven languages

Siamak Barzegar, Brian Davis, Manel Zarrouk, Siegfried Handschuh, and André Freitas. Semr-11: A multi-lingual gold-standard for semantic similarity and relatedness for eleven languages. In LREC, 2018

work page 2018

[3] [3]

Enriching word vectors with subword information

Piotr Bojanowski, Edouard Grave, Armand Joulin, and Tomas Mikolov. Enriching word vectors with subword information. TACL, 2017

work page 2017

[4] [4]

Alternative structures for character-level rnns

Piotr Bojanowski, Armand Joulin, and Tomas Mikolov. Alternative structures for character-level rnns. ICLR, 2015

work page 2015

[5] [5]

A uniﬁed architecture for natural language processing: Deep neural networks with multitask learning

Ronan Collobert and Jason Weston. A uniﬁed architecture for natural language processing: Deep neural networks with multitask learning. In ICML, pages 160–167. ACM, 2008

work page 2008

[6] [6]

Morphological word-embeddings

Ryan Cotterell and Hinrich Schütze. Morphological word-embeddings. In NAACL-HLT, pages 1287–1292, 2015

work page 2015

[7] [7]

Unsupervised discovery of mor- phemes

Mathias Creutz and Krista Lagus. Unsupervised discovery of mor- phemes. In Proceedings of the ACL-02 workshop on Morphological and phonological learning-Volume 6 , pages 21–30. ACL, 2002

work page 2002

[8] [8]

Knet: A general framework for learning word embedding using morphological knowledge

Qing Cui, Bin Gao, Jiang Bian, Siyu Qiu, Hanjun Dai, and Tie-Yan Liu. Knet: A general framework for learning word embedding using morphological knowledge. TOIS, 34(1):4, 2015

work page 2015