• Publications
  • Influence
What do Neural Machine Translation Models Learn about Morphology?
tl;dr
We analyze the representations learned by neural MT models at various levels of granularity and empirically evaluate the quality of the representations for learning morphology through extrinsic part-of-speech and morphological tagging. Expand
  • 180
  • 17
  • Open Access
Evaluating Layers of Representation in Neural Machine Translation on Part-of-Speech and Semantic Tagging Tasks
tl;dr
We train NMT systems on parallel data and use the trained models to extract features for training a classifier on two tasks. Expand
  • 79
  • 6
  • Open Access
Identifying and Controlling Important Neurons in Neural Machine Translation
tl;dr
We develop unsupervised methods for discovering important neurons in NMT models. Expand
  • 48
  • 3
  • Open Access
What Is One Grain of Sand in the Desert? Analyzing Individual Neurons in Deep NLP Models
tl;dr
We address this gap by studying individual dimensions (neurons) in the vector representations learned by end-to-end neural models in NLP tasks. Expand
  • 38
  • 2
  • Open Access
Neural Machine Translation Training in a Multi-Domain Scenario
tl;dr
In this paper, we explore alternative ways to train a neural machine translation system in a multi-domain scenario. Expand
  • 26
  • 2
  • Open Access
Incremental Decoding and Training Methods for Simultaneous Translation in Neural Machine Translation
tl;dr
We address the problem of simultaneous translation by modifying the Neural MT decoder to operate with dynamically built encoder and attention. Expand
  • 22
  • 2
  • Open Access
QCRI $@$ DSL 2016: Spoken Arabic Dialect Identification Using Textual Features
tl;dr
The paper describes the QCRI submissions to the shared task of automatic Arabic dialect classification into 5 Arabic variants, namely Egyptian, Gulf, Levantine, North-African (Maghrebi), and Modern Standard Arabic (MSA). Expand
  • 16
  • 2
  • Open Access
QCRI Machine Translation Systems for IWSLT 16
tl;dr
This paper describes QCRI's machine translation systems for the IWSLT 2016 evaluation campaign for the Arabic->English and English->Arabic tracks. Expand
  • 12
  • 2
  • Open Access
Fighting the COVID-19 Infodemic: Modeling the Perspective of Journalists, Fact-Checkers, Social Media Platforms, Policy Makers, and the Society
tl;dr
We design, annotate, and release a new dataset for fine-grained disinformation analysis that (i) focuses on COVID-19, (ii) combines the perspectives and the interests of journalists, fact-checkers, social media platforms, policy makers, and society as a whole, and (iii) covers both English and Arabic. Expand
  • 7
  • 2
  • Open Access
Understanding and Improving Morphological Learning in the Neural Machine Translation Decoder
tl;dr
We analyze how much morphology an NMT decoder learns, and ii) investigate whether injecting target morphology into the decoder helps it produce better translations. Expand
  • 38
  • 1