TectoMT: Modular NLP Framework

By Martin Popel, Zdenek Zabokrtský. 2010

136 citations recorded by the source corpus.

Subfields: Natural Language Processing, Machine Translation

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

In Machine Translation

Citations in the corpus

Sequence-to-Sequence Generation for Spoken Dialogue via Deep Syntax Trees and Strings (2016, methodology): “We lowercase the data and treat plural -s as separate tokens for generating into strings, and we apply automatic analysis from the Treex NLP toolkit (Popel and ˇZabokrtsk´y, 2010) to obtain deep syntax trees for training tree-based generator setups.”
New Language Pairs in TectoMT (2015, background): “Most TectoMT/Treex rules for the conversion from surface dependencies to deep syntax (t-layer) have been adapted to expect Interset morphological features and HamleDT-style dependencies, which improves their usability for different lan-99 guages.”
Training a Natural Language Generator From Unaligned Data (2015, methodology): “Deep-syntax annotation of sentences in the training set is needed to train the sentence planner, but we assume automatic annotation and reuse an existing deep-syntactic analyzer from the Treex NLP framework (Popel and Žabokrtský, 2010).”
Two-step translation with grammatical post-processing (2011, methodology): “The DEPFIX system was implemented in TectoMT framework (Popel and Žabokrtský, 2010).”
Cross-lingual Coreference Resolution of Pronouns (2014, methodology): “The analysis was carried out within the Treex framework (Popel and ˇZabokrtsk´y, 2010).”
HindEnCorp - Hindi-English and Hindi-only Corpus for Machine Translation (2014, unknown): “For morphological analysis of the English side of the parallel data we employed Morče tagger (Spoustová et al., 2007) available in the Treex platform (Popel and Žabokrtský, 2010) 13 , formerly known as Tec-toMT.”
Adaptation of machine translation for multilingual information retrieval in the medical domain (2014, methodology): “All the data (including training, development, and test sets) was tokenized, tagged for part-of-speech, and lemmatized using the Treex NLP framework [74].”
Findings of the 2016 WMT Shared Task on Cross-lingual Pronoun Prediction (2019, unknown): “2010).”
CzEng 1.6: Enlarged Czech-English Parallel Corpus with Processing Tools Dockered (2016, methodology): “3 Rich Annotation in CzEng 1.6 As in previous versions, CzEng is automatically annotated in the multi-purpose NLP processing framework Treex [6] 11 based on the theory of Functional Generative Description [7].”
Chimera – Three Heads for English-to-Czech Translation (2013, methodology): “included in the Treex platform (Popel and Žabokrtský, 2010).”
Dictionary-based Domain Adaptation of MT Systems without Retraining (2016, methodology): “§ We implemented the methods in the Treex NLP framework of Popel and ˇZabokrtsk´y (2010).”
CUNI in WMT15: Chimera Strikes Again (2015, unknown): “• TectoMT (Popel and Žabokrtský, 2010), a deep-syntactic transfer-based system,”
From Balustrades to Pierre Vinken: Looking for Syntax in Transformer Self-Attentions (2019, methodology): “(Popel and Žabokrtský, 2010; Manning et al., 2014; Forcada et al., 2011), consisting of multiple steps of linguistically motivated analyses, such as partof-speech tagging or syntactic parsing, using explicit intermediate representations (e.”
CUNI-LMU Submissions in WMT2016: Chimera Constrained and Beaten (2016, methodology): “For English, we use Treex (Popel and ˇZabokrtsk´y, 2010) to obtain morphological tags, lemmas and dependency parses of the sentences.”