Multilingual part-of-speech tagging with bidirectional long short-term memory models and auxiliary loss

B Plank, A Søgaard, Y Goldberg - arxiv preprint arxiv:1604.05529, 2016‏ - arxiv.org
Bidirectional long short-term memory (bi-LSTM) networks have recently proven successful
for various NLP sequence modeling tasks, but little is known about their reliance to input …

Many languages, one parser

W Ammar, G Mulcaire, M Ballesteros, C Dyer… - Transactions of the …, 2016‏ - direct.mit.edu
We train one multilingual model for dependency parsing and use it to parse sentences in
several languages. The parsing model uses (i) multilingual word clusters and …

Robust multilingual part-of-speech tagging via adversarial training

M Yasunaga, J Kasai, D Radev - arxiv preprint arxiv:1711.04903, 2017‏ - arxiv.org
Adversarial training (AT) is a powerful regularization method for neural networks, aiming to
achieve robustness to input perturbations. Yet, the specific effects of the robustness obtained …

Multilingual projection for parsing truly low-resource languages

Ž Agić, A Johannsen, B Plank, HM Alonso… - Transactions of the …, 2016‏ - direct.mit.edu
We propose a novel approach to cross-lingual part-of-speech tagging and dependency
parsing for truly low-resource languages. Our annotation projection-based approach yields …

SemEval-2016 Task~ 10: Detecting Minimal Semantic Units and their Meanings (DiMSUM)

N Schneider, D Hovy, A Johannsen… - … Workshop on Semantic …, 2016‏ - research.ed.ac.uk
This task combines the labeling of multiword expressions and supersenses (coarse-grained
classes) in an explicit, yet broad-coverage paradigm for lexical semantics. Nine systems …

Cross-lingual morphological tagging for low-resource languages

J Buys, JA Botha - arxiv preprint arxiv:1606.04279, 2016‏ - arxiv.org
Morphologically rich languages often lack the annotated linguistic resources required to
develop accurate natural language processing tools. We propose models suitable for …

Do LSTMs really work so well for PoS tagging?–A replication study

T Horsmann, T Zesch - Proceedings of the 2017 conference on …, 2017‏ - aclanthology.org
A recent study by Plank et al.(2016) found that LSTM-based PoS taggers considerably
improve over the current state-of-the-art when evaluated on the corpora of the Universal …

Sparse coding of neural word embeddings for multilingual sequence labeling

G Berend - Transactions of the Association for Computational …, 2017‏ - direct.mit.edu
In this paper we propose and carefully evaluate a sequence labeling framework which
solely utilizes sparse indicator features derived from dense distributed word representations …

Surface statistics of an unknown language indicate how to parse it

D Wang, J Eisner - Transactions of the Association for Computational …, 2018‏ - direct.mit.edu
We introduce a novel framework for delexicalized dependency parsing in a new language.
We show that useful features of the target language can be extracted automatically from an …

One parser, many languages

W Ammar, G Mulcaire, M Ballesteros, C Dyer, NA Smith - CoRR, 2016‏ - openreview.net
We train one multilingual model for dependency parsing and use it to parse sentences in
several languages. The parsing model uses (i) multilingual word clusters and …