FBK-UEdin Participation to the WMT13 Quality Estimation Shared Task

39 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.

Findings of the 2013 Workshop on Statistical Machine Translation (2013, methodology): “Information about word alignments is used to extract quantitative (amount and distribution of the alignments) and qualitative (importance of the aligned terms) features under the assumption that alignment information can help tasks where sentencelevel semantic relations need to be identified (Souza ”
Bilingual Embeddings and Word Alignments for Translation Quality Estimation (2016, background): “Many features have been explored by participating systems, including lexical, syntactic, semantic, embedding-based features (Shah et al., 2015), as well as features dependent on any details the particular MT systems may provide (Soricut et al., 2012; Ca-margo de Souza et al., 2013).”
Referential translation machines for predicting semantic similarity (2015, background): “2 Tree features for a CCL parsing output (immediate non-terminals replaced with NP) alignment percentage as in de Souza et al. (2013) and potential BLEU, F 1 , WER (Specia et al. 2014), and PER (Specia et al. 2014) scores for S and T. IBM1 Probability {4, 12} Calculates the translation probability…”