• Publications
  • Influence
Development of a Persian Syntactic Dependency Treebank
The annotation process and linguistic properties of the Persian syntactic dependency treebank, which consists of approximately 30,000 sentences annotated with syntactic roles in addition to morpho-syntactic features, are described.
Yara Parser: A Fast and Accurate Dependency Parser
The Yara Parser is introduced, a fast and accurate open-source dependency parser based on the arc-eager algorithm and beam search that achieves an unlabeled accuracy of 93.32 on the standard WSJ test set which ranks it among the top dependency parsers.
Density-Driven Cross-Lingual Transfer of Dependency Parsers
This work presents a novel method for the crosslingual transfer of dependency parsers that assumes access to parallel translations between the target and one or more source languages, and to supervised parsers in the source language(s).
A Syntactic Valency Lexicon for Persian Verbs : The First Steps towards Persian Dependency Treebank
Valency lexicons are valuable resources for natural language processing. The need for new resources for languages encourages researchers to collect new datasets. One of the most important datasets is
Cross-lingual sentiment transfer with limited resources
This work uses multiple source languages to learn a more robust sentiment transfer model for 16 languages from different language families and shows that it can build a robust transfer system whose performance can in some cases approach that of a supervised system.
Cross-Lingual Transfer of Semantic Roles: From Raw Text to Semantic Roles
We describe a transfer method based on annotation projection to develop a dependency-based semantic role labeling system for languages for which no supervised linguistic information other than
Low-Resource Syntactic Transfer with Unsupervised Source Reordering
It is demonstrated that reordering the source treebanks before training on them for a target language improves the accuracy of languages outside the European language family.
Unsupervised Morphology-Based Vocabulary Expansion
We present a novel way of generating unseen words, which is useful for certain applications such as automatic speech recognition or optical character recognition in low-resource languages. We test
Transferring Semantic Roles Using Translation and Syntactic Information
This paper proposes a transfer method that employs information from source and target syntactic dependencies as well as word alignment density to improve the quality of an iterative bootstrapping method.
ParsiNLU: A Suite of Language Understanding Challenges for Persian
This work introduces ParsiNLU, the first benchmark in Persian language that includes a range of language understanding tasks—reading comprehension, textual entailment, and so on, and presents the first results on state-of-the-art monolingual and multilingual pre-trained language models on this benchmark and compares them with human performance.