Applying Morphology Generation Models to Machine Translation

By Kristina Toutanova, Hisami Suzuki, Achim Ruopp. 2008

162 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

Citations in the corpus

Neural Generation for Czech: Data and Baselines (2019, positive): “This approach is inspired by similar approaches in phrase-based MT (Bojar, 2007; Toutanova et al., 2008; Fraser, 2009) and was developed in parallel to recent similar experiments with two-step neural MT (Nadejde et al., 2017; Tamchyna et al., 2017).”
Modeling Target-Side Inflection in Neural Machine Translation (2017, methodology): “Notably, two-step SMT was proposed to address this problem (Toutanova et al., 2008; Bojar and Kos, 2010; Fraser et al., 2012).”
Evaluating the Morphosyntactic Well-formedness of Generated Texts (2021, background): “In addition to BLEU, chrF and t-BLEU are commonly used to evaluate translation into morphologically-rich languages (Goldwater and McClosky, 2005; Toutanova et al., 2008; Chahuneau et al., 2013; Sennrich et al., 2016).”
Phrase-Based SMT for Finnish with More Data, Better Models and Alternative Alignment and Translation Tools (2016, background): “This approach has been successfully applied to Russian and Arabic (Toutanova et al., 2008) and to German (Fraser et al. (2012), Cap et al. (2014)).”
Morphological Inflection Generation with Hard Monotonic Attention (2016, methodology): “…language and then applying inﬂection generation as a post-processing step is beneﬁcial for phrase-based machine translation (Minkov et al., 2007; Toutanova et al., 2008; Clifton and Sarkar, 2011; Fraser et al., 2012; Chahuneau et al., 2013) and more recently for neural machine translation…”
Investigating the Usefulness of Generalized Word Representations in SMT (2014, methodology): “The idea of translating to stems and then inﬂecting the stems in a separate step has been studied by Toutanova et al. (2008), de Gispert and Mariˆno (2008), Fraser et al. (2012), Chahuneau et al. (2013) and others.”
Morphological Inflection Generation Using Character Sequence to Sequence Learning (2015, methodology): “We plot the character vectors of these Finnish vowels (cf. Figure 5) using t-SNE projection (van der Maaten and Hinton, 2008) and observe that the vowels are correctly grouped with visible transition from the back to the front vowels.”
CimS - The CIS and IMS Joint Submission to WMT 2015 addressing morphological and syntactic differences in English to German SMT (2015, methodology): “While developed for Russian and Arabic morphology, we adapted the approach of Toutanova et al. (2008) to the needs of German in (Fraser et al., 2012).”
Modelling and Optimizing on Syntactic N-Grams for Statistical Machine Translation (2015, background): “, 2012) or independent models for inflection generation (Toutanova et al., 2008; Weller et al., 2013).”
Rich Morphology Generation Using Statistical Machine Translation (2012, background): “Minkov et al. (2007) and Toutanova et al. (2008) model translation and morphology independently for English-Arabic and English-Russian MT.”
On internal language representations in deep learning: an analysis of machine translation and speech recognition (2018, methodology): “1 Another method for handling morphology in machine translation is to use various morphological properties as features, an approach that has been extensively studied in non-neural machine translation [98, 110, 134, 154, 183, 240, 321], and more recently in neural machine translation [153, 302].”
A Hybrid Morpheme-Word Representation for Machine Translation of Morphologically Rich Languages (2010, background): “Research in translating into morphologically rich languages, has attracted interest for languages like Arabic (Badr et al., 2008), Greek (Avramidis and Koehn, 2008), Hungarian (Novák, 2009; Koehn and Haddow, 2009), Russian (Toutanova et al., 2008), and Turkish (Oflazer and El-Kahlout, 2007).”
CimS – The CIS and IMS joint submission to WMT 2014 translating from English into German (2014, methodology): “Re-Inﬂection The two-step translation approach we use was described by e.g. Toutanova et al. (2008) and Jeong et al. (2010), who use a number of morphological and syntactic features derived from both source and target language.”
Using subcategorization knowledge to improve case prediction for translation to German (2013, background): “Toutanova et al. (2008) and Jeong et al. (2010) built translation systems that predict in-ﬂected word forms based on a large array of morphological and syntactic features, obtained from both source and target side.”