CMU-01 at the SIGMORPHON 2019 Shared Task on Crosslinguality and Context in Morphology

Chaudhary, Aditi; Salesky, Elizabeth; Bhat, Gayatri; Mortensen, David R.; Carbonell, Jaime G.; Tsvetkov, Yulia

doi:10.18653/v1/w19-4208

“…Cross-lingual sharing informed by typology has been investigated for, among others, parsing (Naseem et al, 2012;Täckström et al, 2013;Zhang and Barzilay, 2015;de Lhoneux et al, 2018), language modeling Ponti et al, 2019b), machine translation (Daiber et al, 2016;, and morphological inflection (Chaudhary et al, 2019). Many of these approaches use language embeddings with sparse features encoding WALS feature values.…”

Section: Typologically Informed Sharingmentioning

confidence: 99%

SIGTYP 2020 Shared Task: Prediction of Typological Features

Bjerva

¹

,

Salesky²,

Mielke³

et al. 2020

Proceedings of the Second Workshop on Computational Research in Linguistic Typology

Self Cite

View full text Add to dashboard Cite

Typological knowledge bases (KBs) such as WALS (Dryer and Haspelmath, 2013) contain information about linguistic properties of the world's languages. They have been shown to be useful for downstream applications, including cross-lingual transfer learning and linguistic probing. A major drawback hampering broader adoption of typological KBs is that they are sparsely populated, in the sense that most languages only have annotations for some features, and skewed, in that few features have wide coverage. As typological features often correlate with one another, it is possible to predict them and thus automatically populate typological KBs, which is also the focus of this shared task. Overall, the task attracted 8 submissions from 5 teams, out of which the most successful methods make use of such feature correlations. However, our error analysis reveals that even the strongest submitted systems struggle with predicting feature values for languages where few features are known.

show abstract

“…Furthermore, it has proved to be helpful when concatenating word-level and character-level knowledge (Devlin et al, 2019;Peters et al, 2018;Peters et al, 2017). In the morphological analysis task of SIGMORPHON 2019, almost all of the researchers use two levels of word representations to capture more inner-and out-word information (McCarthy et al, 2019;Oh et al, 2019;Chaudhary et al, 2019).…”

Section: Integrating Inner-word and Out-word Featuresmentioning

confidence: 99%

Incorporating Inner-word and Out-word Features for Mongolian Morphological Segmentation

Liu¹,

Su²,

Zhang³

et al. 2020

Proceedings of the 28th International Conference on Computational Linguistics

View full text Add to dashboard Cite

Mongolian morphological segmentation is regarded as a crucial preprocessing step in many Mongolian related NLP applications and has received extensive attention. Recently, end-to-end segmentation approaches with long short-term memory networks (LSTM) have achieved excellent results. However, the inner-word features among characters in the word and the out-word features from context are not well utilized in the segmentation process. In this paper, we propose a neural network incorporating inner-word and out-word features for Mongolian morphological segmentation. The network consists of two encoders and one decoder. The inner-word encoder uses the self-attention mechanisms to capture the inner-word features of the target word. The out-word encoder employs a two layers BiLSTM network to extract out-word features in the sentence. Then, the decoder adopts a multi-head double attention layer to fuse the inner-word features and out-word features and produces the segmentation result. The evaluation experiment compares the proposed network with the baselines and explores the effectiveness of the sub-modules.

show abstract

“…Cross-lingual sharing informed by typology has been investigated for, among others, parsing (Naseem et al, 2012;Täckström et al, 2013;Zhang and Barzilay, 2015;de Lhoneux et al, 2018), language modeling Ponti et al, 2019b), machine translation (Daiber et al, 2016;, and morphological inflection (Chaudhary et al, 2019). Many of these approaches use language embeddings with sparse features encoding WALS feature values.…”

Section: Typologically Informed Sharingmentioning

confidence: 99%

SIGTYP 2020 Shared Task: Prediction of Typological Features

Bjerva¹,

Salesky²,

Mielke³

et al. 2020

Preprint

Self Cite

0

View full text Add to dashboard Cite

Typological knowledge bases (KBs) such as WALS (Dryer and Haspelmath, 2013) contain information about linguistic properties of the world's languages. They have been shown to be useful for downstream applications, including cross-lingual transfer learning and linguistic probing. A major drawback hampering broader adoption of typological KBs is that they are sparsely populated, in the sense that most languages only have annotations for some features, and skewed, in that few features have wide coverage. As typological features often correlate with one another, it is possible to predict them and thus automatically populate typological KBs, which is also the focus of this shared task. Overall, the task attracted 8 submissions from 5 teams, out of which the most successful methods make use of such feature correlations. However, our error analysis reveals that even the strongest submitted systems struggle with predicting feature values for languages where few features are known.

show abstract

CMU-01 at the SIGMORPHON 2019 Shared Task on Crosslinguality and Context in Morphology

Cited by 4 publications

References 21 publications

SIGTYP 2020 Shared Task: Prediction of Typological Features

SIGTYP 2020 Shared Task: Prediction of Typological Features

Incorporating Inner-word and Out-word Features for Mongolian Morphological Segmentation

SIGTYP 2020 Shared Task: Prediction of Typological Features

Contact Info

Product

Resources

About