Successful machine learning systems currently rely on massive amounts of data, which are very effective in hiding some of the shallowness of the learned models. To help train models with more complex and compositional skills, we need challenging data, on which a system is successful only if it detects structure and regularities, that will allow it to generalize. In this paper, we describe a French dataset (BLM-AgrF) for learning the underlying rules of subject-verb agreement in sentences, developed in the BLM framework, a new task inspired by visual IQ tests known as Raven’s Progressive Matrices. In this task, an instance consists of sequences of sentences with specific attributes. To predict the correct answer as the next element of the sequence, a model must correctly detect the generative model used to produce the dataset. We provide details and share a dataset built following this methodology. Two exploratory baselines based on commonly used architectures show that despite the simplicity of the phenomenon, it is a complex problem for deep learning systems.

To develop a system with near-human language capabilities, we need to understand current systems’ generalisation and compositional abilities. We approach this by generating compositional, structured data, inspired from visual intelligence tests, that depend on the problem-solvers being able to disentangle objects and their absolute and relative properties in a sequence of images. We design an analogous task and develop the corresponding datasets that capture specific linguistic phenomena and their properties. Solving each problem instance depends on detecting the relevant linguistic objects and generative rules of the problem. We propose two datasets modelling two linguistic phenomena – subject-verb agreement in French, and verb alternations in English. The datasets can be used to investigate how LLMs encode linguistic objects, such as phrases, their grammatical and semantic properties, such as number or semantic role, and how such information is combined to correctly solve each problem. Specifically generated error types help investigate the behaviour of the system, which important information it is able to detect, and which structures mislead it.

Current NLP models appear to be achieving performance comparable to human capabilities on well-established benchmarks. New benchmarks are now necessary to test deeper layers of understanding of natural languages by these models. Blackbird’s Language Matrices are a recently developed framework that draws inspiration from tests of human analytic intelligence. The BLM task has revealed that successful performances in previously studied linguistic problems do not yet stem from a deep understanding of the generative factors that define these problems. In this study, we define a new BLM task for predicate-argument structure, and develop a structured dataset for its investigation, concentrating on the spray-load verb alternations in English, as a case study. The context sentences include one alternant from the spray-load alternation and the target sentence is the other alternant, to be chosen among a minimally contrastive and adversarial set of answers. We describe the generation process of the dataset and the reasoning behind the generating rules. The dataset aims to facilitate investigations into how verb information is encoded in sentence embeddings and how models generalize to the complex properties of argument structures. Benchmarking experiments conducted on the dataset and qualitative error analysis on the answer set reveal the inherent challenges associated with the problem even for current high-performing representations.

Sentence embeddings induced with various transformer architectures encode much semantic and syntactic information in a distributed manner in a one-dimensional array. We investigate whether specific grammatical information can be accessed in these distributed representations. Using data from a task developed to test rule-like generalizations, our experiments on detecting subject-verb agreement yield several promising results. First, we show that while the usual sentence representations encoded as one-dimensional arrays do not easily support extraction of rule-like regularities, a two-dimensional reshaping of these vectors allows various learning architectures to access such information. Next, we show that various architectures can detect patterns in these two-dimensional reshaped sentence embeddings and successfully learn a model based on smaller amounts of simpler training data, which performs well on more complex test data. This indicates that current sentence embeddings contain information that is regularly distributed, and which can be captured when the embeddings are reshaped into higher dimensional arrays. Our results cast light on representations produced by language models and help move towards developing few-shot learning approaches.

Knowledge graphs, which provide numerous facts in a machine-friendly format, are incomplete. Information that we induce from such graphs – e.g. entity embeddings, relation representations or patterns – will be affected by the imbalance in the information captured in the graph – by biasing representations, or causing us to miss potential patterns. To partially compensate for this situation we describe a method for representing knowledge graphs that capture an intensional representation of the original extensional information. This representation is very compact, and it abstracts away from individual links, allowing us to find better path candidates, as shown by the results of link prediction using this information.

Commonsense knowledge relations are crucial for advanced NLU tasks. We examine the learnability of such relations as represented in ConceptNet, taking into account their specific properties, which can make relation classification difficult: a given concept pair can be linked by multiple relation types, and relations can have multi-word arguments of diverse semantic types. We explore a neural open world multi-label classification approach that focuses on the evaluation of classification accuracy for individual relations. Based on an in-depth study of the specific properties of the ConceptNet resource, we investigate the impact of different relation representations and model variations. Our analysis reveals that the complexity of argument types and relation ambiguity are the most important challenges to address. We design a customized evaluation method to address the incompleteness of the resource that can be expanded in future work.

In this paper, we present experiments that estimate the impact of specific lexical choices of people writing in a second language (L2). In particular, we look at misspelled words that indicate lexical uncertainty on the part of the author, and separate them into three categories: misspelled cognates, “L2-ed” (in our case, anglicized) words, and all other spelling errors. We test the assumption that such errors contain clues about the native language of an essay’s author through the task of native language identification. The results of the experiments show that the information brought by each of these categories is complementary. We also note that while the distribution of such features changes with the proficiency level of the writer, their contribution towards native language identification remains significant at all levels.

We present an analysis of metaphors in news text simplification. Using features that capture general and metaphor specific characteristics, we test whether we can automatically identify which metaphors will be changed or preserved, and whether there are features that have different predictive power for metaphors or literal words. The experiments show that the Age of Acquisition is the most distinctive feature for both metaphors and literal words. Features that capture Imageability and Concreteness are useful when used alone, but within the full set of features they lose their impact. Frequency of use seems to be the best feature to differentiate metaphors that should be changed and those to be preserved.

We introduce a new dataset consisting of natural language interactions annotated with medical family histories, obtained during interactions with a genetic counselor and through crowdsourcing, following a questionnaire created by experts in the domain. We describe the data collection process and the annotations performed by medical professionals, including illness and personal attributes (name, age, gender, family relationships) for the patient and their family members. An initial system that performs argument identification and relation extraction shows promising results – average F-score of 0.87 on complex sentences on the targeted relations.

We induce and visualize a Knowledge Graph over the Regesta Imperii (RI), an important large-scale resource for medieval history research. The RI comprise more than 150,000 digitized abstracts of medieval charters issued by the Roman-German kings and popes distributed over many European locations and a time span of more than 700 years. Our goal is to provide a resource for historians to visualize and query the RI, possibly aiding medieval history research. The resulting medieval graph and visualization tools are shared publicly.

We explore the hypothesis that emotion is one of the dimensions of language that surfaces from the native language into a second language. To check the role of emotions in native language identification (NLI), we model emotion information through polarity and emotion load features, and use document representations using these features to classify the native language of the author. The results indicate that emotion is relevant for NLI, even for high proficiency levels and across topics.

In this paper, we describe experiments designed to explore and evaluate the impact of punctuation marks on the task of native language identification. Punctuation is specific to each language, and is part of the indicators that overtly represent the manner in which each language organizes and conveys information. Our experiments are organized in various set-ups: the usual multi-class classification for individual languages, also considering classification by language groups, across different proficiency levels, topics and even cross-corpus. The results support our hypothesis that punctuation marks are persistent and robust indicators of the native language of the author, which do not diminish in influence even when a high proficiency level in a non-native language is achieved.

We present experiments that show the influence of native language on lexical choice when producing text in another language – in this particular case English. We start from the premise that non-native English speakers will choose lexical items that are close to words in their native language. This leads us to an etymology-based representation of documents written by people whose mother tongue is an Indo-European language. Based on this representation we grow a language family tree, that matches closely the Indo-European language tree.

Detecting aspectual properties of clauses in the form of situation entity types has been shown to depend on a combination of syntactic-semantic and contextual features. We explore this task in a deep-learning framework, where tuned word representations capture lexical, syntactic and semantic features. We introduce an attention mechanism that pinpoints relevant context not only for the current instance, but also for the larger context. Apart from implicitly capturing task relevant features, the advantage of our neural model is that it avoids the need to reproduce linguistic features for other languages and is thus more easily transferable. We present experiments for English and German that achieve competitive performance. We present a novel take on modeling and exploiting genre information and showcase the adaptation of our system from one language to another.

In this paper, we explore spelling errors as a source of information for detecting the native language of a writer, a previously under-explored area. We note that character n-grams from misspelled words are very indicative of the native language of the author. In combination with other lexical features, spelling error features lead to 1.2% improvement in accuracy on classifying texts in the TOEFL11 corpus by the author’s native language, compared to systems participating in the NLI shared task.

Every non-trivial text describes interactions and relations between people, institutions, activities, events and so on. What we know about the world consists in large part of such relations, and that knowledge contributes to the understanding of what texts refer to. Newly found relations can in turn become part of this knowledge that is stored for future use.To grasp a text’s semantic content, an automatic system must be able to recognize relations in texts and reason about them. This may be done by applying and updating previously acquired knowledge. We focus here in particular on semantic relations which describe the interactions among nouns and compact noun phrases, and we present such relations from both a theoretical and a practical perspective. The theoretical exploration sketches the historical path which has brought us to the contemporary view and interpretation of semantic relations. We discuss a wide range of relation inventories proposed by linguists and by language processing people. Such inventories vary by domain, granularity and suitability for downstream applications.On the practical side, we investigate the recognition and acquisition of relations from texts. In a look at supervised learning methods, we present available datasets, the variety of features which can describe relation instances, and learning algorithms found appropriate for the task. Next, we present weakly supervised and unsupervised learning methods of acquiring relations from large corpora with little or no previously annotated data. We show how enduring the bootstrapping algorithm based on seed examples or patterns has proved to be, and how it has been adapted to tackle Web-scale text collections. We also show a few machine learning techniques which can perform fast and reliable relation extraction by taking advantage of data redundancy and variability.

In this paper we present the mapping between WordNet domains and WordNet topics, and the emergent Wikipedia categories. This mapping leads to a coarse alignment between WordNet and Wikipedia, useful for producing domain-specific and multilingual corpora. Multilinguality is achieved through the cross-language links between Wikipedia categories. Research in word-sense disambiguation has shown that within a specific domain, relevant words have restricted senses. The multilingual, and comparable, domain-specific corpora we produce have the potential to enhance research in word-sense disambiguation and terminology extraction in different languages, which could enhance the performance of various NLP tasks.

This paper describes the derivation of distributional semantic representations for open class words relative to a concept inventory, and of concepts relative to open class words through grammatical relations extracted from Wikipedia articles. The concept inventory comes from WikiNet, a large-scale concept network derived from Wikipedia. The distinctive feature of these representations are their relation to a concept network, through which we can compute selectional preferences of open-class words relative to general concepts. The resource thus derived provides a meaning representation that complements the relational representation captured in the concept network. It covers English open-class words, but the concept base is language independent. The resource can be extended to other languages, with the use of language specific dependency parsers. Good results in metonymy resolution show the resource's potential use for NLP applications.

This paper describes a multi-lingual large-scale concept network obtained automatically by mining for concepts and relations and exploiting a variety of sources of knowledge from Wikipedia. Concepts and their lexicalizations are extracted from Wikipedia pages, in particular from article titles, hyperlinks, disambiguation pages and cross-language links. Relations are extracted from the category and page network, from the category names, from infoboxes and the body of the articles. The resulting network has two main components: (i) a central, language independent index of concepts, which serves to keep track of the concepts' lexicalizations both within a language and across languages, and to separate linguistic expressions of concepts from the relations in which they are involved (concepts themselves are represented as numeric IDs); (ii) a large network built on the basis of the relations extracted, represented as relations between concepts (more specifically, the numeric IDs). The various stages of obtaining the network were separately evaluated, and the results show a qualitative resource.

This paper presents the process of acquiring a large, domain independent, taxonomy from the German Wikipedia. We build upon a previously implemented platform that extracts a semantic network and taxonomy from the English version of the Wikipedia. We describe two accomplishments of our work: the semantic network for the German language in which isa links are identified and annotated, and an expansion of the platform for easy adaptation for a new language. We identify the platform’s strengths and shortcomings, which stem from the scarcity of free processing resources for languages other than English. We show that the taxonomy induction process is highly reliable - evaluated against the German version of WordNet, GermaNet, the resource obtained shows an accuracy of 83.34%.