Standard test sets for supervised learning evaluate in-distribution generalization. Unfortunately, when a dataset has systematic gaps (e.g., annotation artifacts), these evaluations are misleading: a model can learn simple decision rules that perform well on the test set but do not capture the abilities a dataset is intended to test. We propose a more rigorous annotation paradigm for NLP that helps to close systematic gaps in the test data. In particular, after a dataset is constructed, we recommend that the dataset authors manually perturb the test instances in small but meaningful ways that (typically) change the gold label, creating contrast sets. Contrast sets provide a local view of a model's decision boundary, which can be used to more accurately evaluate a model's true linguistic capabilities. We demonstrate the efficacy of contrast sets by creating them for 10 diverse NLP datasets (e.g., DROP reading comprehension, UD parsing, and IMDb sentiment analysis). Although our contrast sets are not explicitly adversarial, model performance is significantly lower on them than on the original test sets-up to 25% in some cases. We release our contrast sets as new evaluation benchmarks and encourage future dataset construction efforts to follow similar annotation processes.

Universal dependencies (UD) is a framework for morphosyntactic annotation of human language, which to date has been used to create treebanks for more than 100 languages. In this article, we outline the linguistic theory of the UD framework, which draws on a long tradition of typologically oriented grammatical theories. Grammatical relations between words are centrally used to explain how predicate–argument structures are encoded morphosyntactically in different languages while morphological features and part-of-speech classes give the properties of words. We argue that this theory is a good basis for cross-linguistically consistent annotation of typologically diverse languages in a way that supports computational natural language understanding as well as broader linguistic studies.

In standard NLP pipelines, morphological analysis and disambiguation (MA&D) precedes syntactic and semantic downstream tasks. However, for languages with complex and ambiguous word-internal structure, known as morphologically rich languages (MRLs), it has been hypothesized that syntactic context may be crucial for accurate MA&D, and vice versa. In this work we empirically confirm this hypothesis for Modern Hebrew, an MRL with complex morphology and severe word-level ambiguity, in a novel transition-based framework. Specifically, we propose a joint morphosyntactic transition-based framework which formally unifies two distinct transition systems, morphological and syntactic, into a single transition-based system with joint training and joint inference. We empirically show that MA&D results obtained in the joint settings outperform MA&D results obtained by the respective standalone components, and that end-to-end parsing results obtained by our joint system present a new state of the art for Hebrew dependency parsing.

International audienceParsing is a key task in natural language processing. It involves predicting, for each natural language sentence, an abstract representation of the grammatical entities in the sentence and the relations between these entities. This representation provides an interface to compositional semantics and to the notions of "who did what to whom." The last two decades have seen great advances in parsing English, leading to major leaps also in the performance of applications that use parsers as part of their backbone, such as systems for information extraction, sentiment analysis, text summarization, and machine translation. Attempts to replicate the success of parsing English for other languages have often yielded unsatisfactory results. In particular, parsing languages with complex word structure and flexible word order has been shown to require non-trivial adaptation. This special issue reports on methods that successfully address the challenges involved in parsing a range of morphologically rich languages (MRLs). This introduction characterizes MRLs, describes the challenges in parsing MRLs, and outlines the contributions of the articles in the special issue. These contributions present up-to-date research efforts that address parsing in varied, cross-lingual settings. They show that parsing MRLs addresses challenges that transcend particular representational and algorithmic choices

Current parsing models are not immediately applicable for languages that exhibit strong interaction between morphology and syntax, e.g., Modern Hebrew (MH), Arabic and other Semitic languages. This work represents a first attempt at modeling morphological-syntactic interaction in a generative probabilistic framework to allow for MH parsing. We show that morphological information selected in tandem with syntactic categories is instrumental for parsing Semitic languages. We further show that redundant morphological information helps syntactic disambiguation.

We present a framework for interfacing a PCFG parser with lexical information from an external resource following a different tagging scheme than the treebank. This is achieved by defining a stochastic mapping layer between the two resources. Lexical probabilities for rare events are estimated in a semi-supervised manner from a lexicon and large unannotated corpora. We show that this solution greatly enhances the performance of an unlexicalized Hebrew PCFG parser, resulting in state-of-the-art Hebrew parsing results both when a segmentation oracle is assumed, and in a real-word parsing scenario of parsing unsegmented tokens.

State-of-the-art statistical parsing models applied to free word-order languages tend to underperform compared to, e.g., parsing English. Constituency-based models often fail to capture generalizations that cannot be stated in structural terms, and dependency-based models employ a 'single-head' assumption that often breaks in the face of multiple exponence. In this paper we suggest that the position of a constituent is a form manifestation of its grammatical function, one among various possible means of realization. We develop the Relational-Realizational approach to parsing in which we untangle the projection of grammatical functions and their means of realization to allow for phrase-structure variability and morphological-syntactic interaction. We empirically demonstrate the application of our approach to parsing Modern Hebrew, obtaining 7% error reduction from previously reported results.

No abstract