Lluís Màrquez

Lluís Màrquez

Auto-generated profile, based on publicly available information

Lluís Màrquez

Arabic Language Technologies Research Group Qatar Computing Research Institute HBKU

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2017 in Computer Speech & Language
Verified
Journal Article
Closed access
FGFrancisco Guzmán
Francisco Guzmán
SJShafiq Joty
Shafiq Joty
LMLluís Màrquez
Lluís Màrquez
PNPreslav Nakov
Preslav Nakov

1 December 2017 in Computational Linguistics
Verified
Journal Article
Open access
SJShafiq Joty
Shafiq Joty
FGFrancisco Guzmán
Francisco Guzmán
LMLluís Màrquez
Lluís Màrquez
PNPreslav Nakov
Preslav Nakov
In this article, we explore the potential of using sentence-level discourse structure for machine translation evaluation. We first design discourse-aware similarity measures, which use all-subtree kernels to compare discourse parse trees in accordance with the Rhetorical Structure Theory (RST). Then, we show that a simple linear combination with these measures can help improve various existing machine translation evaluation metrics regarding correlation with human judgments both at the segment level and at the system level. This suggests that discourse information is complementary to the information used by many of the existing evaluation metrics, and thus it could be taken into account when developing richer evaluation metrics, such as the WMT-14 winning combined metric DiscoTKparty. We also provide a detailed analysis of the relevance of various discourse elements and relations from the RST parse trees for machine translation evaluation. In particular, we show that (i) all aspects of the RST tree are relevant, (ii) nuclearity is more useful than relation type, and (iii) the similarity of the translation RST tree to the reference RST tree is positively correlated with translation quality.

16 September 2014 in Machine Translation
Verified
Journal Article
Closed access
PDF file
GLGorka Labaka
Gorka Labaka
CECristina España-Bonet
Cristina España-Bonet
LMLluís Màrquez
Lluís Màrquez
KSKepa Sarasola
Kepa Sarasola

28 September 2015 in Journal of Artificial Intelligence Research
Verified
Journal Article
Closed access
PDF file
LFLluís Formiga
Lluís Formiga
ABAlberto Barrón-Cedeño
Alberto Barrón-Cedeño
LMLluís Màrquez
Lluís Màrquez
CHCarlos A. Henríquez
Carlos A. Henríquez
JMJosé B. Mariño
José B. Mariño


23 January 2018 in arXiv
Verified
Preprint
Open access
PDF file
YBYonatan Belinkov
Yonatan Belinkov
LMLluís Màrquez
Lluís Màrquez
HSHassan Sajjad
Hassan Sajjad
NDNadir Durrani
Nadir Durrani
FDFahim Dalvi
Fahim Dalvi
JGJames Glass
James Glass
While neural machine translation (NMT) models provide improved translation quality in an elegant, end-to-end framework, it is less clear what they learn about language. Recent work has started evaluating the quality of vector representations learned by NMT models on morphological and syntactic tasks. In this paper, we investigate the representations learned at different layers of NMT encoders. We train NMT systems on parallel data and use the trained models to extract features for training a classifier on two tasks: part-of-speech and semantic tagging. We then measure the performance of the classifier as a proxy to the quality of the original NMT model for the given task. Our quantitative analysis yields interesting insights regarding representation learning in NMT models. For instance, we find that higher layers are better at learning semantics while lower layers tend to be better for part-of-speech tagging. We also observe little effect of the target language on source-side representations, especially with higher quality NMT models.