Goran Glavaš


2020

pdf bib
On the Limitations of Cross-lingual Encoders as Exposed by Reference-Free Machine Translation Evaluation
Wei Zhao | Goran Glavaš | Maxime Peyrard | Yang Gao | Robert West | Steffen Eger
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics

Evaluation of cross-lingual encoders is usually performed either via zero-shot cross-lingual transfer in supervised downstream tasks or via unsupervised cross-lingual textual similarity. In this paper, we concern ourselves with reference-free machine translation (MT) evaluation where we directly compare source texts to (sometimes low-quality) system translations, which represents a natural adversarial setup for multilingual encoders. Reference-free evaluation holds the promise of web-scale comparison of MT systems. We systematically investigate a range of metrics based on state-of-the-art cross-lingual semantic representations obtained with pretrained M-BERT and LASER. We find that they perform poorly as semantic encoders for reference-free MT evaluation and identify their two key limitations, namely, (a) a semantic mismatch between representations of mutual translations and, more prominently, (b) the inability to punish “translationese”, i.e., low-quality literal translations. We propose two partial remedies: (1) post-hoc re-alignment of the vector spaces and (2) coupling of semantic-similarity based metrics with target-side language modeling. In segment-level MT evaluation, our best metric surpasses reference-based BLEU by 5.7 correlation points.

pdf bib
Classification-Based Self-Learning for Weakly Supervised Bilingual Lexicon Induction
Mladen Karan | Ivan Vulić | Anna Korhonen | Goran Glavaš
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics

Effective projection-based cross-lingual word embedding (CLWE) induction critically relies on the iterative self-learning procedure. It gradually expands the initial small seed dictionary to learn improved cross-lingual mappings. In this work, we present ClassyMap, a classification-based approach to self-learning, yielding a more robust and a more effective induction of projection-based CLWEs. Unlike prior self-learning methods, our approach allows for integration of diverse features into the iterative process. We show the benefits of ClassyMap for bilingual lexicon induction: we report consistent improvements in a weakly supervised setup (500 seed translation pairs) on a benchmark with 28 language pairs.

pdf bib
Non-Linear Instance-Based Cross-Lingual Mapping for Non-Isomorphic Embedding Spaces
Goran Glavaš | Ivan Vulić
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics

We present InstaMap, an instance-based method for learning projection-based cross-lingual word embeddings. Unlike prior work, it deviates from learning a single global linear projection. InstaMap is a non-parametric model that learns a non-linear projection by iteratively: (1) finding a globally optimal rotation of the source embedding space relying on the Kabsch algorithm, and then (2) moving each point along an instance-specific translation vector estimated from the translation vectors of the point’s nearest neighbours in the training dictionary. We report performance gains with InstaMap over four representative state-of-the-art projection-based models on bilingual lexicon induction across a set of 28 diverse language pairs. We note prominent improvements, especially for more distant language pairs (i.e., languages with non-isomorphic monolingual spaces).

pdf bib
SemEval-2020 Task 2: Predicting Multilingual and Cross-Lingual (Graded) Lexical Entailment
Goran Glavaš | Ivan Vulić | Anna Korhonen | Simone Paolo Ponzetto
Proceedings of the Fourteenth Workshop on Semantic Evaluation

Lexical entailment (LE) is a fundamental asymmetric lexico-semantic relation, supporting the hierarchies in lexical resources (e.g., WordNet, ConceptNet) and applications like natural language inference and taxonomy induction. Multilingual and cross-lingual NLP applications warrant models for LE detection that go beyond language boundaries. As part of SemEval 2020, we carried out a shared task (Task 2) on multilingual and cross-lingual LE. The shared task spans three dimensions: (1) monolingual vs. cross-lingual LE, (2) binary vs. graded LE, and (3) a set of 6 diverse languages (and 15 corresponding language pairs). We offered two different evaluation tracks: (a) Dist: for unsupervised, fully distributional models that capture LE solely on the basis of unannotated corpora, and (b) Any: for externally informed models, allowed to leverage any resources, including lexico-semantic networks (e.g., WordNet or BabelNet). In the Any track, we recieved runs that push state-of-the-art across all languages and language pairs, for both binary LE detection and graded LE prediction.

pdf bib
Proceedings of SustaiNLP: Workshop on Simple and Efficient Natural Language Processing
Nafise Sadat Moosavi | Angela Fan | Vered Shwartz | Goran Glavaš | Shafiq Joty | Alex Wang | Thomas Wolf
Proceedings of SustaiNLP: Workshop on Simple and Efficient Natural Language Processing

pdf bib
Common Sense or World Knowledge? Investigating Adapter-Based Knowledge Injection into Pretrained Transformers
Anne Lauscher | Olga Majewska | Leonardo F. R. Ribeiro | Iryna Gurevych | Nikolai Rozanov | Goran Glavaš
Proceedings of Deep Learning Inside Out (DeeLIO): The First Workshop on Knowledge Extraction and Integration for Deep Learning Architectures

Following the major success of neural language models (LMs) such as BERT or GPT-2 on a variety of language understanding tasks, recent work focused on injecting (structured) knowledge from external resources into these models. While on the one hand, joint pre-training (i.e., training from scratch, adding objectives based on external knowledge to the primary LM objective) may be prohibitively computationally expensive, post-hoc fine-tuning on external knowledge, on the other hand, may lead to the catastrophic forgetting of distributional knowledge. In this work, we investigate models for complementing the distributional knowledge of BERT with conceptual knowledge from ConceptNet and its corresponding Open Mind Common Sense (OMCS) corpus, respectively, using adapter training. While overall results on the GLUE benchmark paint an inconclusive picture, a deeper analysis reveals that our adapter-based models substantially outperform BERT (up to 15-20 performance points) on inference tasks that require the type of conceptual knowledge explicitly present in ConceptNet and OMCS. We also open source all our experiments and relevant code under: https://github.com/wluper/retrograph.

pdf bib
AraWEAT: Multidimensional Analysis of Biases in Arabic Word Embeddings
Anne Lauscher | Rafik Takieddin | Simone Paolo Ponzetto | Goran Glavaš
Proceedings of the Fifth Arabic Natural Language Processing Workshop

Recent work has shown that distributional word vector spaces often encode human biases like sexism or racism. In this work, we conduct an extensive analysis of biases in Arabic word embeddings by applying a range of recently introduced bias tests on a variety of embedding spaces induced from corpora in Arabic. We measure the presence of biases across several dimensions, namely: embedding models (Skip-Gram, CBOW, and FastText) and vector sizes, types of text (encyclopedic text, and news vs. user-generated content), dialects (Egyptian Arabic vs. Modern Standard Arabic), and time (diachronic analyses over corpora from different time periods). Our analysis yields several interesting findings, e.g., that implicit gender bias in embeddings trained on Arabic news corpora steadily increases over time (between 2007 and 2017). We make the Arabic bias specifications (AraWEAT) publicly available.

pdf bib
Specializing Unsupervised Pretraining Models for Word-Level Semantic Similarity
Anne Lauscher | Ivan Vulić | Edoardo Maria Ponti | Anna Korhonen | Goran Glavaš
Proceedings of the 28th International Conference on Computational Linguistics

Unsupervised pretraining models have been shown to facilitate a wide range of downstream NLP applications. These models, however, retain some of the limitations of traditional static word embeddings. In particular, they encode only the distributional knowledge available in raw text corpora, incorporated through language modeling objectives. In this work, we complement such distributional knowledge with external lexical knowledge, that is, we integrate the discrete knowledge on word-level semantic similarity into pretraining. To this end, we generalize the standard BERT model to a multi-task learning setting where we couple BERT’s masked language modeling and next sentence prediction objectives with an auxiliary task of binary word relation classification. Our experiments suggest that our “Lexically Informed” BERT (LIBERT), specialized for the word-level semantic similarity, yields better performance than the lexically blind “vanilla” BERT on several language understanding tasks. Concretely, LIBERT outperforms BERT in 9 out of 10 tasks of the GLUE benchmark and is on a par with BERT in the remaining one. Moreover, we show consistent gains on 3 benchmarks for lexical simplification, a task where knowledge about word-level semantic similarity is paramount, as well as large gains on lexical reasoning probes.

pdf bib
Towards Instance-Level Parser Selection for Cross-Lingual Transfer of Dependency Parsers
Robert Litschko | Ivan Vulić | Željko Agić | Goran Glavaš
Proceedings of the 28th International Conference on Computational Linguistics

Current methods of cross-lingual parser transfer focus on predicting the best parser for a low-resource target language globally, that is, “at treebank level”. In this work, we propose and argue for a novel cross-lingual transfer paradigm: instance-level parser selection (ILPS), and present a proof-of-concept study focused on instance-level selection in the framework of delexicalized parser transfer. Our work is motivated by an empirical observation that different source parsers are the best choice for different Universal POS-sequences (i.e., UPOS sentences) in the target language. We then propose to predict the best parser at the instance level. To this end, we train a supervised regression model, based on the Transformer architecture, to predict parser accuracies for individual POS-sequences. We compare ILPS against two strong single-best parser selection baselines (SBPS): (1) a model that compares POS n-gram distributions between the source and target languages (KL) and (2) a model that selects the source based on the similarity between manually created language vectors encoding syntactic properties of languages (L2V). The results from our extensive evaluation, coupling 42 source parsers and 20 diverse low-resource test languages, show that ILPS outperforms KL and L2V on 13/20 and 14/20 test languages, respectively. Further, we show that by predicting the best parser “at treebank level” (SBPS), using the aggregation of predictions from our instance-level model, we outperform the same baselines on 17/20 and 16/20 test languages.

pdf bib
XHate-999: Analyzing and Detecting Abusive Language Across Domains and Languages
Goran Glavaš | Mladen Karan | Ivan Vulić
Proceedings of the 28th International Conference on Computational Linguistics

We present XHate-999, a multi-domain and multilingual evaluation data set for abusive language detection. By aligning test instances across six typologically diverse languages, XHate-999 for the first time allows for disentanglement of the domain transfer and language transfer effects in abusive language detection. We conduct a series of domain- and language-transfer experiments with state-of-the-art monolingual and multilingual transformer models, setting strong baseline results and profiling XHate-999 as a comprehensive evaluation resource for abusive language detection. Finally, we show that domain- and language-adaption, via intermediate masked language modeling on abusive corpora in the target language, can lead to substantially improved abusive language detection in the target language in the zero-shot transfer setups.

pdf bib
XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning
Edoardo Maria Ponti | Goran Glavaš | Olga Majewska | Qianchu Liu | Ivan Vulić | Anna Korhonen
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

In order to simulate human language capacity, natural language processing systems must be able to reason about the dynamics of everyday situations, including their possible causes and effects. Moreover, they should be able to generalise the acquired world knowledge to new languages, modulo cultural differences. Advances in machine reasoning and cross-lingual transfer depend on the availability of challenging evaluation benchmarks. Motivated by both demands, we introduce Cross-lingual Choice of Plausible Alternatives (XCOPA), a typologically diverse multilingual dataset for causal commonsense reasoning in 11 languages, which includes resource-poor languages like Eastern Apurímac Quechua and Haitian Creole. We evaluate a range of state-of-the-art models on this novel dataset, revealing that the performance of current methods based on multilingual pretraining and zero-shot fine-tuning falls short compared to translation-based transfer. Finally, we propose strategies to adapt multilingual models to out-of-sample resource-lean languages where only a small corpus or a bilingual dictionary is available, and report substantial improvements over the random baseline. The XCOPA dataset is freely available at github.com/cambridgeltl/xcopa.

pdf bib
From Zero to Hero: On the Limitations of Zero-Shot Language Transfer with Multilingual Transformers
Anne Lauscher | Vinit Ravishankar | Ivan Vulić | Goran Glavaš
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

Massively multilingual transformers (MMTs) pretrained via language modeling (e.g., mBERT, XLM-R) have become a default paradigm for zero-shot language transfer in NLP, offering unmatched transfer performance. Current evaluations, however, verify their efficacy in transfers (a) to languages with sufficiently large pretraining corpora, and (b) between close languages. In this work, we analyze the limitations of downstream language transfer with MMTs, showing that, much like cross-lingual word embeddings, they are substantially less effective in resource-lean scenarios and for distant languages. Our experiments, encompassing three lower-level tasks (POS tagging, dependency parsing, NER) and two high-level tasks (NLI, QA), empirically correlate transfer performance with linguistic proximity between source and target languages, but also with the size of target language corpora used in MMT pretraining. Most importantly, we demonstrate that the inexpensive few-shot transfer (i.e., additional fine-tuning on a few target-language instances) is surprisingly effective across the board, warranting more research efforts reaching beyond the limiting zero-shot conditions.

pdf bib
Probing Pretrained Language Models for Lexical Semantics
Ivan Vulić | Edoardo Maria Ponti | Robert Litschko | Goran Glavaš | Anna Korhonen
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

The success of large pretrained language models (LMs) such as BERT and RoBERTa has sparked interest in probing their representations, in order to unveil what types of knowledge they implicitly capture. While prior research focused on morphosyntactic, semantic, and world knowledge, it remains unclear to which extent LMs also derive lexical type-level knowledge from words in context. In this work, we present a systematic empirical analysis across six typologically diverse languages and five different lexical tasks, addressing the following questions: 1) How do different lexical knowledge extraction strategies (monolingual versus multilingual source LM, out-of-context versus in-context encoding, inclusion of special tokens, and layer-wise averaging) impact performance? How consistent are the observed effects across tasks and languages? 2) Is lexical knowledge stored in few parameters, or is it scattered throughout the network? 3) How do these representations fare against traditional static word vectors in lexical tasks 4) Does the lexical information emerging from independently trained monolingual LMs display latent similarities? Our main results indicate patterns and best practices that hold universally, but also point to prominent variations across languages and tasks. Moreover, we validate the claim that lower Transformer layers carry more type-level lexical knowledge, but also show that this knowledge is distributed across multiple layers.

pdf bib
Improving Bilingual Lexicon Induction with Unsupervised Post-Processing of Monolingual Word Vector Spaces
Ivan Vulić | Anna Korhonen | Goran Glavaš
Proceedings of the 5th Workshop on Representation Learning for NLP

Work on projection-based induction of cross-lingual word embedding spaces (CLWEs) predominantly focuses on the improvement of the projection (i.e., mapping) mechanisms. In this work, in contrast, we show that a simple method for post-processing monolingual embedding spaces facilitates learning of the cross-lingual alignment and, in turn, substantially improves bilingual lexicon induction (BLI). The post-processing method we examine is grounded in the generalisation of first- and second-order monolingual similarities to the nth-order similarity. By post-processing monolingual spaces before the cross-lingual alignment, the method can be coupled with any projection-based method for inducing CLWE spaces. We demonstrate the effectiveness of this simple monolingual post-processing across a set of 15 typologically diverse languages (i.e., 15*14 BLI setups), and in combination with two different projection methods.

2019

pdf bib
Cross-lingual Semantic Specialization via Lexical Relation Induction
Edoardo Maria Ponti | Ivan Vulić | Goran Glavaš | Roi Reichart | Anna Korhonen
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)

Semantic specialization integrates structured linguistic knowledge from external resources (such as lexical relations in WordNet) into pretrained distributional vectors in the form of constraints. However, this technique cannot be leveraged in many languages, because their structured external resources are typically incomplete or non-existent. To bridge this gap, we propose a novel method that transfers specialization from a resource-rich source language (English) to virtually any target language. Our specialization transfer comprises two crucial steps: 1) Inducing noisy constraints in the target language through automatic word translation; and 2) Filtering the noisy constraints via a state-of-the-art relation prediction model trained on the source language constraints. This allows us to specialize any set of distributional vectors in the target language with the refined constraints. We prove the effectiveness of our method through intrinsic word similarity evaluation in 8 languages, and with 3 downstream tasks in 5 languages: lexical simplification, dialog state tracking, and semantic textual similarity. The gains over the previous state-of-art specialization methods are substantial and consistent across languages. Our results also suggest that the transfer method is effective even for lexically distant source-target language pairs. Finally, as a by-product, our method produces lists of WordNet-style lexical relations in resource-poor languages.

pdf bib
Do We Really Need Fully Unsupervised Cross-Lingual Embeddings?
Ivan Vulić | Goran Glavaš | Roi Reichart | Anna Korhonen
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)

Recent efforts in cross-lingual word embedding (CLWE) learning have predominantly focused on fully unsupervised approaches that project monolingual embeddings into a shared cross-lingual space without any cross-lingual signal. The lack of any supervision makes such approaches conceptually attractive. Yet, their only core difference from (weakly) supervised projection-based CLWE methods is in the way they obtain a seed dictionary used to initialize an iterative self-learning procedure. The fully unsupervised methods have arguably become more robust, and their primary use case is CLWE induction for pairs of resource-poor and distant languages. In this paper, we question the ability of even the most robust unsupervised CLWE approaches to induce meaningful CLWEs in these more challenging settings. A series of bilingual lexicon induction (BLI) experiments with 15 diverse languages (210 language pairs) show that fully unsupervised CLWE methods still fail for a large number of language pairs (e.g., they yield zero BLI performance for 87/210 pairs). Even when they succeed, they never surpass the performance of weakly supervised methods (seeded with 500-1,000 translation pairs) using the same self-learning procedure in any BLI setup, and the gaps are often substantial. These findings call for revisiting the main motivations behind fully unsupervised CLWE methods.

pdf bib
SEAGLE: A Platform for Comparative Evaluation of Semantic Encoders for Information Retrieval
Fabian David Schmidt | Markus Dietsche | Simone Paolo Ponzetto | Goran Glavaš
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP): System Demonstrations

We introduce Seagle, a platform for comparative evaluation of semantic text encoding models on information retrieval (IR) tasks. Seagle implements (1) word embedding aggregators, which represent texts as algebraic aggregations of pretrained word embeddings and (2) pretrained semantic encoders, and allows for their comparative evaluation on arbitrary (monolingual and cross-lingual) IR collections. We benchmark Seagle’s models on monolingual document retrieval and cross-lingual sentence retrieval. Seagle functionality can be exploited via an easy-to-use web interface and its modular backend (micro-service architecture) can easily be extended with additional semantic search models.

pdf bib
Proceedings of the Thirteenth Workshop on Graph-Based Methods for Natural Language Processing (TextGraphs-13)
Dmitry Ustalov | Swapna Somasundaran | Peter Jansen | Goran Glavaš | Martin Riedl | Mihai Surdeanu | Michalis Vazirgiannis
Proceedings of the Thirteenth Workshop on Graph-Based Methods for Natural Language Processing (TextGraphs-13)

pdf bib
How to (Properly) Evaluate Cross-Lingual Word Embeddings: On Strong Baselines, Comparative Analyses, and Some Misconceptions
Goran Glavaš | Robert Litschko | Sebastian Ruder | Ivan Vulić
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

Cross-lingual word embeddings (CLEs) facilitate cross-lingual transfer of NLP models. Despite their ubiquitous downstream usage, increasingly popular projection-based CLE models are almost exclusively evaluated on bilingual lexicon induction (BLI). Even the BLI evaluations vary greatly, hindering our ability to correctly interpret performance and properties of different CLE models. In this work, we take the first step towards a comprehensive evaluation of CLE models: we thoroughly evaluate both supervised and unsupervised CLE models, for a large number of language pairs, on BLI and three downstream tasks, providing new insights concerning the ability of cutting-edge CLE models to support cross-lingual NLP. We empirically demonstrate that the performance of CLE models largely depends on the task at hand and that optimizing CLE models for BLI may hurt downstream performance. We indicate the most robust supervised and unsupervised CLE models and emphasize the need to reassess simple baselines, which still display competitive performance across the board. We hope our work catalyzes further research on CLE evaluation and model analysis.

pdf bib
Generalized Tuning of Distributional Word Vectors for Monolingual and Cross-Lingual Lexical Entailment
Goran Glavaš | Ivan Vulić
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

Lexical entailment (LE; also known as hyponymy-hypernymy or is-a relation) is a core asymmetric lexical relation that supports tasks like taxonomy induction and text generation. In this work, we propose a simple and effective method for fine-tuning distributional word vectors for LE. Our Generalized Lexical ENtailment model (GLEN) is decoupled from the word embedding model and applicable to any distributional vector space. Yet – unlike existing retrofitting models – it captures a general specialization function allowing for LE-tuning of the entire distributional space and not only the vectors of words seen in lexical constraints. Coupled with a multilingual embedding space, GLEN seamlessly enables cross-lingual LE detection. We demonstrate the effectiveness of GLEN in graded LE and report large improvements (over 20% in accuracy) over state-of-the-art in cross-lingual LE detection.

pdf bib
Multilingual and Cross-Lingual Graded Lexical Entailment
Ivan Vulić | Simone Paolo Ponzetto | Goran Glavaš
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

Grounded in cognitive linguistics, graded lexical entailment (GR-LE) is concerned with fine-grained assertions regarding the directional hierarchical relationships between concepts on a continuous scale. In this paper, we present the first work on cross-lingual generalisation of GR-LE relation. Starting from HyperLex, the only available GR-LE dataset in English, we construct new monolingual GR-LE datasets for three other languages, and combine those to create a set of six cross-lingual GR-LE datasets termed CL-HYPERLEX. We next present a novel method dubbed CLEAR (Cross-Lingual Lexical Entailment Attract-Repel) for effectively capturing graded (and binary) LE, both monolingually in different languages as well as across languages (i.e., on CL-HYPERLEX). Coupled with a bilingual dictionary, CLEAR leverages taxonomic LE knowledge in a resource-rich language (e.g., English) and propagates it to other languages. Supported by cross-lingual LE transfer, CLEAR sets competitive baseline performance on three new monolingual GR-LE datasets and six cross-lingual GR-LE datasets. In addition, we show that CLEAR outperforms current state-of-the-art on binary cross-lingual LE detection by a wide margin for diverse language pairs.

pdf bib
Computational Analysis of Political Texts: Bridging Research Efforts Across Communities
Goran Glavaš | Federico Nanni | Simone Paolo Ponzetto
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: Tutorial Abstracts

In the last twenty years, political scientists started adopting and developing natural language processing (NLP) methods more actively in order to exploit text as an additional source of data in their analyses. Over the last decade the usage of computational methods for analysis of political texts has drastically expanded in scope, allowing for a sustained growth of the text-as-data community in political science. In political science, NLP methods have been extensively used for a number of analyses types and tasks, including inferring policy position of actors from textual evidence, detecting topics in political texts, and analyzing stylistic aspects of political texts (e.g., assessing the role of language ambiguity in framing the political agenda). Just like in numerous other domains, much of the work on computational analysis of political texts has been enabled and facilitated by the development of resources such as, the topically coded electoral programmes (e.g., the Manifesto Corpus) or topically coded legislative texts (e.g., the Comparative Agenda Project). Political scientists created resources and used available NLP methods to process textual data largely in isolation from the NLP community. At the same time, NLP researchers addressed closely related tasks such as election prediction, ideology classification, and stance detection. In other words, these two communities have been largely agnostic of one another, with NLP researchers mostly unaware of interesting applications in political science and political scientists not applying cutting-edge NLP methodology to their problems. The main goal of this tutorial is to systematize and analyze the body of research work on political texts from both communities. We aim to provide a gentle, all-round introduction to methods and tasks related to computational analysis of political texts. Our vision is to bring the two research communities closer to each other and contribute to faster and more significant developments in this interdisciplinary research area.

pdf bib
Specializing Distributional Vectors of All Words for Lexical Entailment
Aishwarya Kamath | Jonas Pfeiffer | Edoardo Maria Ponti | Goran Glavaš | Ivan Vulić
Proceedings of the 4th Workshop on Representation Learning for NLP (RepL4NLP-2019)

Semantic specialization methods fine-tune distributional word vectors using lexical knowledge from external resources (e.g. WordNet) to accentuate a particular relation between words. However, such post-processing methods suffer from limited coverage as they affect only vectors of words seen in the external resources. We present the first post-processing method that specializes vectors of all vocabulary words – including those unseen in the resources – for the asymmetric relation of lexical entailment (LE) (i.e., hyponymy-hypernymy relation). Leveraging a partially LE-specialized distributional space, our POSTLE (i.e., post-specialization for LE) model learns an explicit global specialization function, allowing for specialization of vectors of unseen words, as well as word vectors from other languages via cross-lingual transfer. We capture the function as a deep feed-forward neural network: its objective re-scales vector norms to reflect the concept hierarchy while simultaneously attracting hyponymy-hypernymy pairs to better reflect semantic similarity. An extended model variant augments the basic architecture with an adversarial discriminator. We demonstrate the usefulness and versatility of POSTLE models with different input distributional spaces in different scenarios (monolingual LE and zero-shot cross-lingual LE transfer) and tasks (binary and graded LE). We report consistent gains over state-of-the-art LE-specialization methods, and successfully LE-specialize word vectors for languages without any external lexical knowledge.

pdf bib
SenZi: A Sentiment Analysis Lexicon for the Latinised Arabic (Arabizi)
Taha Tobaili | Miriam Fernandez | Harith Alani | Sanaa Sharafeddine | Hazem Hajj | Goran Glavaš
Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP 2019)

Arabizi is an informal written form of dialectal Arabic transcribed in Latin alphanumeric characters. It has a proven popularity on chat platforms and social media, yet it suffers from a severe lack of natural language processing (NLP) resources. As such, texts written in Arabizi are often disregarded in sentiment analysis tasks for Arabic. In this paper we describe the creation of a sentiment lexicon for Arabizi that was enriched with word embeddings. The result is a new Arabizi lexicon consisting of 11.3K positive and 13.3K negative words. We evaluated this lexicon by classifying the sentiment of Arabizi tweets achieving an F1-score of 0.72. We provide a detailed error analysis to present the challenges that impact the sentiment analysis of Arabizi.

pdf bib
Are We Consistently Biased? Multidimensional Analysis of Biases in Distributional Word Vectors
Anne Lauscher | Goran Glavaš
Proceedings of the Eighth Joint Conference on Lexical and Computational Semantics (*SEM 2019)

Word embeddings have recently been shown to reflect many of the pronounced societal biases (e.g., gender bias or racial bias). Existing studies are, however, limited in scope and do not investigate the consistency of biases across relevant dimensions like embedding models, types of texts, and different languages. In this work, we present a systematic study of biases encoded in distributional word vector spaces: we analyze how consistent the bias effects are across languages, corpora, and embedding models. Furthermore, we analyze the cross-lingual biases encoded in bilingual embedding spaces, indicative of the effects of bias transfer encompassed in cross-lingual transfer of NLP models. Our study yields some unexpected findings, e.g., that biases can be emphasized or downplayed by different embedding models or that user-generated content may be less biased than encyclopedic text. We hope our work catalyzes bias research in NLP and informs the development of bias reduction techniques.

2018

pdf bib
Post-Specialisation: Retrofitting Vectors of Words Unseen in Lexical Resources
Ivan Vulić | Goran Glavaš | Nikola Mrkšić | Anna Korhonen
Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers)

Word vector specialisation (also known as retrofitting) is a portable, light-weight approach to fine-tuning arbitrary distributional word vector spaces by injecting external knowledge from rich lexical resources such as WordNet. By design, these post-processing methods only update the vectors of words occurring in external lexicons, leaving the representations of all unseen words intact. In this paper, we show that constraint-driven vector space specialisation can be extended to unseen words. We propose a novel post-specialisation method that: a) preserves the useful linguistic knowledge for seen words; while b) propagating this external signal to unseen words in order to improve their vector representations as well. Our post-specialisation approach explicits a non-linear specialisation function in the form of a deep neural network by learning to predict specialised vectors from their original distributional counterparts. The learned function is then used to specialise vectors of unseen words. This approach, applicable to any post-processing model, yields considerable gains over the initial specialisation models both in intrinsic word similarity tasks, and in two downstream tasks: dialogue state tracking and lexical text simplification. The positive effects persist across three languages, demonstrating the importance of specialising the full vocabulary of distributional word vector spaces.

pdf bib
Discriminating between Lexico-Semantic Relations with the Specialization Tensor Model
Goran Glavaš | Ivan Vulić
Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers)

We present a simple and effective feed-forward neural architecture for discriminating between lexico-semantic relations (synonymy, antonymy, hypernymy, and meronymy). Our Specialization Tensor Model (STM) simultaneously produces multiple different specializations of input distributional word vectors, tailored for predicting lexico-semantic relations for word pairs. STM outperforms more complex state-of-the-art architectures on two benchmark datasets and exhibits stable performance across languages. We also show that, if coupled with a bilingual distributional space, the proposed model can transfer the prediction of lexico-semantic relations to a resource-lean target language without any training data.

pdf bib
Adversarial Propagation and Zero-Shot Cross-Lingual Transfer of Word Vector Specialization
Edoardo Maria Ponti | Ivan Vulić | Goran Glavaš | Nikola Mrkšić | Anna Korhonen
Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing

Semantic specialization is a process of fine-tuning pre-trained distributional word vectors using external lexical knowledge (e.g., WordNet) to accentuate a particular semantic relation in the specialized vector space. While post-processing specialization methods are applicable to arbitrary distributional vectors, they are limited to updating only the vectors of words occurring in external lexicons (i.e., seen words), leaving the vectors of all other words unchanged. We propose a novel approach to specializing the full distributional vocabulary. Our adversarial post-specialization method propagates the external lexical knowledge to the full distributional space. We exploit words seen in the resources as training examples for learning a global specialization function. This function is learned by combining a standard L2-distance loss with a adversarial loss: the adversarial component produces more realistic output vectors. We show the effectiveness and robustness of the proposed method across three languages and on three tasks: word similarity, dialog state tracking, and lexical simplification. We report consistent improvements over distributional word vectors and vectors specialized by other state-of-the-art specialization frameworks. Finally, we also propose a cross-lingual transfer method for zero-shot specialization which successfully specializes a full target distributional space without any lexical knowledge in the target language and without any bilingual data.

pdf bib
Investigating the Role of Argumentation in the Rhetorical Analysis of Scientific Publications with Neural Multi-Task Learning Models
Anne Lauscher | Goran Glavaš | Simone Paolo Ponzetto | Kai Eckert
Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing

Exponential growth in the number of scientific publications yields the need for effective automatic analysis of rhetorical aspects of scientific writing. Acknowledging the argumentative nature of scientific text, in this work we investigate the link between the argumentative structure of scientific publications and rhetorical aspects such as discourse categories or citation contexts. To this end, we (1) augment a corpus of scientific publications annotated with four layers of rhetoric annotations with argumentation annotations and (2) investigate neural multi-task learning architectures combining argument extraction with a set of rhetorical classification tasks. By coupling rhetorical classifiers with the extraction of argumentative components in a joint multi-task learning setting, we obtain significant performance gains for different rhetorical analysis tasks.

pdf bib
Explicit Retrofitting of Distributional Word Vectors
Goran Glavaš | Ivan Vulić
Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

Semantic specialization of distributional word vectors, referred to as retrofitting, is a process of fine-tuning word vectors using external lexical knowledge in order to better embed some semantic relation. Existing retrofitting models integrate linguistic constraints directly into learning objectives and, consequently, specialize only the vectors of words from the constraints. In this work, in contrast, we transform external lexico-semantic relations into training examples which we use to learn an explicit retrofitting model (ER). The ER model allows us to learn a global specialization function and specialize the vectors of words unobserved in the training data as well. We report large gains over original distributional vector spaces in (1) intrinsic word similarity evaluation and on (2) two downstream tasks − lexical simplification and dialog state tracking. Finally, we also successfully specialize vector spaces of new languages (i.e., unseen in the training data) by coupling ER with shared multilingual distributional vector spaces.

pdf bib
Proceedings of the Twelfth Workshop on Graph-Based Methods for Natural Language Processing (TextGraphs-12)
Goran Glavaš | Swapna Somasundaran | Martin Riedl | Eduard Hovy
Proceedings of the Twelfth Workshop on Graph-Based Methods for Natural Language Processing (TextGraphs-12)

pdf bib
ArguminSci: A Tool for Analyzing Argumentation and Rhetorical Aspects in Scientific Writing
Anne Lauscher | Goran Glavaš | Kai Eckert
Proceedings of the 5th Workshop on Argument Mining

Argumentation is arguably one of the central features of scientific language. We present ArguminSci, an easy-to-use tool that analyzes argumentation and other rhetorical aspects of scientific writing, which we collectively dub scitorics. The main aspect we focus on is the fine-grained argumentative analysis of scientific text through identification of argument components. The functionality of ArguminSci is accessible via three interfaces: as a command line tool, via a RESTful application programming interface, and as a web application.

pdf bib
An Argument-Annotated Corpus of Scientific Publications
Anne Lauscher | Goran Glavaš | Simone Paolo Ponzetto
Proceedings of the 5th Workshop on Argument Mining

Argumentation is an essential feature of scientific language. We present an annotation study resulting in a corpus of scientific publications annotated with argumentative components and relations. The argumentative annotations have been added to the existing Dr. Inventor Corpus, already annotated for four other rhetorical aspects. We analyze the annotated argumentative structures and investigate the relations between argumentation and other rhetorical aspects of scientific writing, such as discourse roles and citation contexts.

2017

pdf bib
Improving Neural Knowledge Base Completion with Cross-Lingual Projections
Patrick Klein | Simone Paolo Ponzetto | Goran Glavaš
Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers

In this paper we present a cross-lingual extension of a neural tensor network model for knowledge base completion. We exploit multilingual synsets from BabelNet to translate English triples to other languages and then augment the reference knowledge base with cross-lingual triples. We project monolingual embeddings of different languages to a shared multilingual space and use them for network initialization (i.e., as initial concept embeddings). We then train the network with triples from the cross-lingually augmented knowledge base. Results on WordNet link prediction show that leveraging cross-lingual information yields significant gains over exploiting only monolingual triples.

pdf bib
Unsupervised Cross-Lingual Scaling of Political Texts
Goran Glavaš | Federico Nanni | Simone Paolo Ponzetto
Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers

Political text scaling aims to linearly order parties and politicians across political dimensions (e.g., left-to-right ideology) based on textual content (e.g., politician speeches or party manifestos). Existing models scale texts based on relative word usage and cannot be used for cross-lingual analyses. Additionally, there is little quantitative evidence that the output of these models correlates with common political dimensions like left-to-right orientation. Experimental results show that the semantically-informed scaling models better predict the party positions than the existing word-based models in two different political dimensions. Furthermore, the proposed models exhibit no drop in performance in the cross-lingual compared to monolingual setting.

pdf bib
Two Layers of Annotation for Representing Event Mentions in News Stories
Maria Pia di Buono | Martin Tutek | Jan Šnajder | Goran Glavaš | Bojana Dalbelo Bašić | Nataša Milić-Frayling
Proceedings of the 11th Linguistic Annotation Workshop

In this paper, we describe our preliminary study on annotating event mention as a part of our research on high-precision news event extraction models. To this end, we propose a two-layer annotation scheme, designed to separately capture the functional and conceptual aspects of event mentions. We hypothesize that the precision of models can be improved by modeling and extracting separately the different aspects of news events, and then combining the extracted information by leveraging the complementarities of the models. In addition, we carry out a preliminary annotation using the proposed scheme and analyze the annotation quality in terms of inter-annotator agreement.

pdf bib
Proceedings of TextGraphs-11: the Workshop on Graph-based Methods for Natural Language Processing
Martin Riedl | Swapna Somasundaran | Goran Glavaš | Eduard Hovy
Proceedings of TextGraphs-11: the Workshop on Graph-based Methods for Natural Language Processing

pdf bib
Cross-Lingual Classification of Topics in Political Texts
Goran Glavaš | Federico Nanni | Simone Paolo Ponzetto
Proceedings of the Second Workshop on NLP and Computational Social Science

In this paper, we propose an approach for cross-lingual topical coding of sentences from electoral manifestos of political parties in different languages. To this end, we exploit continuous semantic text representations and induce a joint multilingual semantic vector spaces to enable supervised learning using manually-coded sentences across different languages. Our experimental results show that classifiers trained on multilingual data yield performance boosts over monolingual topic classification.

pdf bib
Predicting News Values from Headline Text and Emotions
Maria Pia di Buono | Jan Šnajder | Bojana Dalbelo Bašić | Goran Glavaš | Martin Tutek | Natasa Milic-Frayling
Proceedings of the 2017 EMNLP Workshop: Natural Language Processing meets Journalism

We present a preliminary study on predicting news values from headline text and emotions. We perform a multivariate analysis on a dataset manually annotated with news values and emotions, discovering interesting correlations among them. We then train two competitive machine learning models – an SVM and a CNN – to predict news values from headline text and emotions as features. We find that, while both models yield a satisfactory performance, some news values are more difficult to detect than others, while some profit more from including emotion information.

pdf bib
If Sentences Could See: Investigating Visual Information for Semantic Textual Similarity
Goran Glavaš | Ivan Vulić | Simone Paolo Ponzetto
IWCS 2017 - 12th International Conference on Computational Semantics - Long papers

pdf bib
Dual Tensor Model for Detecting Asymmetric Lexico-Semantic Relations
Goran Glavaš | Simone Paolo Ponzetto
Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

Detection of lexico-semantic relations is one of the central tasks of computational semantics. Although some fundamental relations (e.g., hypernymy) are asymmetric, most existing models account for asymmetry only implicitly and use the same concept representations to support detection of symmetric and asymmetric relations alike. In this work, we propose the Dual Tensor model, a neural architecture with which we explicitly model the asymmetry and capture the translation between unspecialized and specialized word embeddings via a pair of tensors. Although our Dual Tensor model needs only unspecialized embeddings as input, our experiments on hypernymy and meronymy detection suggest that it can outperform more complex and resource-intensive models. We further demonstrate that the model can account for polysemy and that it exhibits stable performance across languages.

2016

pdf bib
Analysis of Policy Agendas: Lessons Learned from Automatic Topic Classification of Croatian Political Texts
Mladen Karan | Jan Šnajder | Daniela Širinić | Goran Glavaš
Proceedings of the 10th SIGHUM Workshop on Language Technology for Cultural Heritage, Social Sciences, and Humanities

pdf bib
Spanish NER with Word Representations and Conditional Random Fields
Jenny Linet Copara Zea | Jose Eduardo Ochoa Luna | Camilo Thorne | Goran Glavaš
Proceedings of the Sixth Named Entity Workshop

pdf bib
Unsupervised Text Segmentation Using Semantic Relatedness Graphs
Goran Glavaš | Federico Nanni | Simone Paolo Ponzetto
Proceedings of the Fifth Joint Conference on Lexical and Computational Semantics

2015

pdf bib
Resolving Entity Coreference in Croatian with a Constrained Mention-Pair Model
Goran Glavaš | Jan Šnajder
The 5th Workshop on Balto-Slavic Natural Language Processing

pdf bib
Simplifying Lexical Simplification: Do We Need Simplified Corpora?
Goran Glavaš | Sanja Štajner
Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)

pdf bib
TKLBLIIR: Detecting Twitter Paraphrases with TweetingJay
Mladen Karan | Goran Glavaš | Jan Šnajder | Bojana Dalbelo Bašić | Ivan Vulić | Marie-Francine Moens
Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015)

pdf bib
TAKELAB: Medical Information Extraction and Linking with MINERAL
Goran Glavaš
Proceedings of the 9th International Workshop on Semantic Evaluation (SemEval 2015)

2014

pdf bib
Constructing Coherent Event Hierarchies from News Stories
Goran Glavaš | Jan Šnajder
Proceedings of TextGraphs-9: the workshop on Graph-based Methods for Natural Language Processing

pdf bib
HiEve: A Corpus for Extracting Event Hierarchies from News Stories
Goran Glavaš | Jan Šnajder | Marie-Francine Moens | Parisa Kordjamshidi
Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)

In news stories, event mentions denote real-world events of different spatial and temporal granularity. Narratives in news stories typically describe some real-world event of coarse spatial and temporal granularity along with its subevents. In this work, we present HiEve, a corpus for recognizing relations of spatiotemporal containment between events. In HiEve, the narratives are represented as hierarchies of events based on relations of spatiotemporal containment (i.e., superevent―subevent relations). We describe the process of manual annotation of HiEve. Furthermore, we build a supervised classifier for recognizing spatiotemporal containment between events to serve as a baseline for future research. Preliminary experimental results are encouraging, with classifier performance reaching 58% F1-score, only 11% less than the inter annotator agreement.

2013

pdf bib
Aspect-Oriented Opinion Mining from User Reviews in Croatian
Goran Glavaš | Damir Korenčić | Jan Šnajder
Proceedings of the 4th Biennial International Workshop on Balto-Slavic Natural Language Processing

pdf bib
Event-Centered Information Retrieval Using Kernels on Event Graphs
Goran Glavaš | Jan Šnajder
Proceedings of TextGraphs-8 Graph-based Methods for Natural Language Processing

pdf bib
Recognizing Identical Events with Graph Kernels
Goran Glavaš | Jan Šnajder
Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)

pdf bib
Event-Centered Simplification of News Stories
Goran Glavaš | Sanja Štajner
Proceedings of the Student Research Workshop associated with RANLP 2013

2012

pdf bib
TakeLab: Systems for Measuring Semantic Text Similarity
Frane Šarić | Goran Glavaš | Mladen Karan | Jan Šnajder | Bojana Dalbelo Bašić
*SEM 2012: The First Joint Conference on Lexical and Computational Semantics – Volume 1: Proceedings of the main conference and the shared task, and Volume 2: Proceedings of the Sixth International Workshop on Semantic Evaluation (SemEval 2012)

pdf bib
Experiments on Hybrid Corpus-Based Sentiment Lexicon Acquisition
Goran Glavaš | Jan Šnajder | Bojana Dalbelo Bašić
Proceedings of the Workshop on Innovative Hybrid Approaches to the Processing of Textual Data