Generative Large Language Models (LLMs), such as GPT-3, have become increasingly effective and versatile in natural language processing (NLP) tasks. One such task is Lexical Simplification, where state-of-the-art methods involve complex, multi-step processes which can use both deep learning and non-deep learning processes. LLaMA, an LLM with full research access, holds unique potential for the adaption of the entire LS pipeline. This paper details the process of fine-tuning LLaMA to create LSLlama, which performs comparably to previous LS baseline models LSBert and UniHD.

Due to having knowledge of only basic vocabulary, many people cannot understand up-to-date written information and thus make informed decisions and fully participate in the society. We propose LeSS, a modular lexical simplification architecture that outperforms state-of-the-art lexical simplification systems for Spanish. In addition to its state-of-the-art performance, LeSS is computationally light, using much less disk space, CPU and GPU, and having faster loading and execution time than the transformer-based lexical simplification models which are predominant in the field.

BSL-Hansard is a novel open source and multimodal resource composed by combining Sign Language video data in BSL and English text from the official transcription of British parliamentary sessions. This paper describes the method followed to compile BSL-Hansard including time alignment of text using the MAUS (Schiel, 2015) segmentation system, gives some statistics about this dataset, and suggests experiments. These primarily include end-to-end Sign Language-to-text translation, but is also relevant for broader machine translation, and speech and language processing tasks.

This paper examines the use of manually part-of-speech tagged sign language gloss data in the Text2Gloss and Gloss2Text translation tasks, as well as running an LSTM-based sequence labelling model on the same glosses for automatic part-of-speech tagging. We find that a combination of tag-enhanced glosses and pretraining the neural model positively impacts performance in the translation tasks. The results of the tagging task are limited, but provide a methodological framework for further research into tagging sign language gloss data.

SignON (https://signon-project.eu/) is a Horizon 2020 project, running from 2021 until the end of 2023, which addresses the lack of technology and services for the automatic translation between sign languages (SLs) and spoken languages, through an inclusive, human-centric solution, hence contributing to the repertoire of communication media for deaf, hard of hearing (DHH) and hearing individuals. In this paper, we present an update of the status of the project, describing the approaches developed to address the challenges and peculiarities of SL machine translation (SLMT).

Sign Languages (SLs) are the primary means of communication for at least half a million people in Europe alone. However, the development of SL recognition and translation tools is slowed down by a series of obstacles concerning resource scarcity and standardization issues in the available data. The former challenge relates to the volume of data available for machine learning as well as the time required to collect and process new data. The latter obstacle is linked to the variety of the data, i.e., annotation formats are not unified and vary amongst different resources. The available data formats are often not suitable for machine learning, obstructing the provision of automatic tools based on neural models. In the present paper, we give an overview of these challenges by comparing various SL corpora and SL machine learning datasets. Furthermore, we propose a framework to address the lack of standardization at format level, unify the available resources and facilitate SL research for different languages. Our framework takes ELAN files as inputs and returns textual and visual data ready to train SL recognition and translation models. We present a proof of concept, training neural translation models on the data produced by the proposed framework.

Lexical Simplification is the process of reducing the lexical complexity of a text by replacing difficult words with easier to read (or understand) expressions while preserving the original information and meaning. In this paper we introduce ALEXSIS, a new dataset for this task, and we use ALEXSIS to benchmark Lexical Simplification systems in Spanish. The paper describes the evaluation of three kind of approaches to Lexical Simplification, a thesaurus-based approach, a single transformers-based approach, and a combination of transformers. We also report state of the art results on a previous Lexical Simplification dataset for Spanish.

This paper presents a series of experiments on translating between spoken Spanish and Spanish Sign Language glosses (LSE), including enriching Neural Machine Translation (NMT) systems with linguistic features, and creating synthetic data to pretrain and later on finetune a neural translation model. We found evidence that pretraining over a large corpus of LSE synthetic data aligned to Spanish sentences could markedly improve the performance of the translation models.

This study presents a lexical simplification (LS) methodology for foreign language (FL) learning purposes, a barely explored area of automatic text simplification (TS). The method, targeted at Spanish as a foreign language (SFL), includes a customised complex word identification (CWI) classifier and generates substitutions based on masked language modelling. Performance is calculated on a custom dataset by means of a new, pedagogically-oriented evaluation. With 43% of the top simplifications being found suitable, the method shows potential for simplifying sentences to be used in FL learning activities. The evaluation also suggests that, though still crucial, meaning preservation is not always a prerequisite for successful LS. To arrive at grammatically correct and more idiomatic simplifications, future research could study the integration of association measures based on co-occurrence data.

Fine-tuning Transformer-based approaches have recently shown exciting results on sentence simplification task. However, so far, no research has applied similar approaches to the Lexical Simplification (LS) task. In this paper, we present ConLS, a Controllable Lexical Simplification system fine-tuned with T5 (a Transformer-based model pre-trained with a BERT-style approach and several other tasks). The evaluation results on three datasets (LexMTurk, BenchLS, and NNSeval) have shown that our model performs comparable to LSBert (the current state-of-the-art) and even outperforms it in some cases. We also conducted a detailed comparison on the effectiveness of control tokens to give a clear view of how each token contributes to the model.

We report findings of the TSAR-2022 shared task on multilingual lexical simplification, organized as part of the Workshop on Text Simplification, Accessibility, and Readability TSAR-2022 held in conjunction with EMNLP 2022. The task called the Natural Language Processing research community to contribute with methods to advance the state of the art in multilingual lexical simplification for English, Portuguese, and Spanish. A total of 14 teams submitted the results of their lexical simplification systems for the provided test data. Results of the shared task indicate new benchmarks in Lexical Simplification with English lexical simplification quantitative results noticeably higher than those obtained for Spanish and (Brazilian) Portuguese.

Identification de mots et passages difficiles dans les documents médicaux en français. L’objectif de la simplification automatique des textes consiste à fournir une nouvelle version de documents qui devient plus facile à comprendre pour une population donnée ou plus facile à traiter par d’autres applications du TAL. Cependant, avant d’effectuer la simplification, il est important de savoir ce qu’il faut simplifier exactement dans les documents. En effet, même dans les documents techniques et spécialisés, il n’est pas nécessaire de tout simplifier mais juste les segments qui présentent des difficultés de compréhension. Il s’agit typiquement de la tâche d’identification de mots complexes : effectuer le diagnostic de difficulté d’un document donné pour y détecter les mots et passages complexes. Nous proposons de travail sur l’identification de mots et passages complexes dans les documents biomédicaux en français.

Évaluation de la simplification automatique de textes : où nous en sommes et vers où devonsnous aller. L’objectif de la simplification automatique de textes consiste à adapter le contenu de documents afin de les rendre plus faciles à comprendre par une population donnée ou bien pour améliorer les performances d’autres tâches TAL, comme le résumé automatique ou extraction d’information. Les étapes principales de la simplification automatique de textes sont plutôt bien définies et étudiées dans les travaux existants, alors que l’évaluation de la simplification reste sous-étudiée. En effet, contrairement à d’autres tâches de TAL, comme la recherche et extraction d’information, la structuration de terminologie ou les questions-réponses, qui s’attendent à avoir des résultats factuels et consensuels, il est difficile de définir un résultat standard de la simplification. Le processus de simplification est très subjectif et souvent non consensuel parce qu’il est lourdement basé sur les connaissances propres des personnes. Ainsi, plusieurs facteurs sont impliqués dans le processus de simplification et son évaluation. Dans ce papier, nous présentons et discutons quelques uns de ces facteurs : le rôle de l’utilisateur final, les données de référence, le domaine des documents source et les mesures d’évaluation.

Sharing datasets and benchmarks has been crucial for rapidly improving Natural Language Processing models and systems. Documenting datasets’ characteristics (and any modification introduced over time) is equally important to avoid confusion and make comparisons reliable. Here, we describe the case of BigPatent, a dataset for patent summarization that exists in at least two rather different versions under the same name. While previous literature has not clearly distinguished among versions, their differences do not only lay on a surface level but also modify the dataset’s core nature and, thus, the complexity of the summarization task. While this paper describes a specific case, we aim to shed light on new challenges that might emerge in resource sharing and advocate for comprehensive documentation of datasets and models.

This paper is a description of our participation in the Multi-document Summarization for Literature Review (MSLR) Shared Task, in which we explore summarization models to create an automatic review of scientific results. Rather than maximizing the metrics using expensive computational models, we placed ourselves in a situation of scarce computational resources and explore the limits of a base sequence to sequence models (thus with a limited input length) to the task. Although we explore methods to feed the abstractive model with salient sentences only (using a first extractive step), we find the results still need some improvements.

Recently, a large pre-trained language model called T5 (A Unified Text-to-Text Transfer Transformer) has achieved state-of-the-art performance in many NLP tasks. However, no study has been found using this pre-trained model on Text Simplification. Therefore in this paper, we explore the use of T5 fine-tuning on Text Simplification combining with a controllable mechanism to regulate the system outputs that can help generate adapted text for different target audiences. Our experiments show that our model achieves remarkable results with gains of between +0.69 and +1.41 over the current state-of-the-art (BART+ACCESS). We argue that using a pre-trained model such as T5, trained on several tasks with large amounts of data, can help improve Text Simplification.

It is well-established that the preferred mode of communication of the deaf and hard of hearing (DHH) community are Sign Languages (SLs), but they are considered low resource languages where natural language processing technologies are of concern. In this paper we study the problem of text to SL gloss Machine Translation (MT) using Transformer-based architectures. Despite the significant advances of MT for spoken languages in the recent couple of decades, MT is in its infancy when it comes to SLs. We enrich a Transformer-based architecture aggregating syntactic information extracted from a dependency parser to word-embeddings. We test our model on a well-known dataset showing that the syntax-aware model obtains performance gains in terms of MT evaluation metrics.

This paper presents the participation of the LaSTUS/TALN team at TRAC-2020 Trolling, Aggression and Cyberbullying shared task. The aim of the task is to determine whether a given text is aggressive and contains misogynistic content. Our approach is based on a bidirectional Long Short Term Memory network (bi-LSTM). Our system performed well at sub-task A, aggression detection; however underachieved at sub-task B, misogyny detection.

Related work sections or literature reviews are an essential part of every scientific article being crucial for paper reviewing and assessment. The automatic generation of related work sections can be considered an instance of the multi-document summarization problem. In order to allow the study of this specific problem, we have developed a manually annotated, machine readable data-set of related work sections, cited papers (e.g. references) and sentences, together with an additional layer of papers citing the references. We additionally present experiments on the identification of cited sentences, using as input citation contexts. The corpus alongside the gold standard are made available for use by the scientific community.

In this work we propose to leverage resources available with discourse-level annotations to facilitate the identification of argumentative components and relations in scientific texts, which has been recognized as a particularly challenging task. In particular, we implement and evaluate a transfer learning approach in which contextualized representations learned from discourse parsing tasks are used as input of argument mining models. As a pilot application, we explore the feasibility of using automatically identified argumentative components and relations to predict the acceptance of papers in computer science venues. In order to conduct our experiments, we propose an annotation scheme for argumentative units and relations and use it to enrich an existing corpus with an argumentation layer.

We present a bidirectional Long-Short Term Memory network for identifying offensive language in Twitter. Our system has been developed in the context of the SemEval 2019 Task 6 which comprises three different sub-tasks, namely A: Offensive Language Detection, B: Categorization of Offensive Language, C: Offensive Language Target Identification. We used a pre-trained Word Embeddings in tweet data, including information about emojis and hashtags. Our approach achieves good performance in the three sub-tasks.

This paper presents the participation of the LaSTUS/TALN team in the Complex Word Identification (CWI) Shared Task 2018 in the English monolingual track . The purpose of the task was to determine if a word in a given sentence can be judged as complex or not by a certain target audience. For the English track, task organizers provided a training and a development datasets of 27,299 and 3,328 words respectively together with the sentence in which each word occurs. The words were judged as complex or not by 20 human evaluators; ten of whom are natives. We submitted two systems: one system modeled each word to evaluate as a numeric vector populated with a set of lexical, semantic and contextual features while the other system relies on a word embedding representation and a distance metric. We trained two separate classifiers to automatically decide if each word is complex or not. We submitted six runs, two for each of the three subsets of the English monolingual CWI track.

This paper describes the results of the first Shared Task on Multilingual Emoji Prediction, organized as part of SemEval 2018. Given the text of a tweet, the task consists of predicting the most likely emoji to be used along such tweet. Two subtasks were proposed, one for English and one for Spanish, and participants were allowed to submit a system run to one or both subtasks. In total, 49 teams participated to the English subtask and 22 teams submitted a system run to the Spanish subtask. Evaluation was carried out emoji-wise, and the final ranking was based on macro F-Score. Data and further information about this task can be found at https://competitions.codalab.org/competitions/17344.

This paper describes the SemEval 2018 Shared Task on Hypernym Discovery. We put forward this task as a complementary benchmark for modeling hypernymy, a problem which has traditionally been cast as a binary classification task, taking a pair of candidate words as input. Instead, our reformulated task is defined as follows: given an input term, retrieve (or discover) its suitable hypernyms from a target corpus. We proposed five different subtasks covering three languages (English, Spanish, and Italian), and two specific domains of knowledge in English (Medical and Music). Participants were allowed to compete in any or all of the subtasks. Overall, a total of 11 teams participated, with a total of 39 different systems submitted through all subtasks. Data, results and further information about the task can be found at https://competitions.codalab.org/competitions/17119.

Human language has evolved towards newer forms of communication such as social media, where emojis (i.e., ideograms bearing a visual meaning) play a key role. While there is an increasing body of work aimed at the computational modeling of emoji semantics, there is currently little understanding about what makes a computational model represent or predict a given emoji in a certain way. In this paper we propose a label-wise attention mechanism with which we attempt to better understand the nuances underlying emoji prediction. In addition to advantages in terms of interpretability, we show that our proposed architecture improves over standard baselines in emoji prediction, and does particularly well when predicting infrequent emojis.

Emojis are small images that are commonly included in social media text messages. The combination of visual and textual content in the same message builds up a modern way of communication, that automatic systems are not used to deal with. In this paper we extend recent advances in emoji prediction by putting forward a multimodal approach that is able to predict emojis in Instagram posts. Instagram posts are composed of pictures together with texts which sometimes include emojis. We show that these emojis can be predicted by using the text, but also using the picture. Our main finding is that incorporating the two synergistic modalities, in a combined model, improves accuracy in an emoji prediction task. This result demonstrates that these two modalities (text and images) encode different information on the use of emojis and therefore can complement each other.

Videogame streaming platforms have become a paramount example of noisy user-generated text. These are websites where gaming is broadcasted, and allows interaction with viewers via integrated chatrooms. Probably the best known platform of this kind is Twitch, which has more than 100 million monthly viewers. Despite these numbers, and unlike other platforms featuring short messages (e.g. Twitter), Twitch has not received much attention from the Natural Language Processing community. In this paper we aim at bridging this gap by proposing two important tasks specific to the Twitch platform, namely (1) Emote prediction; and (2) Trolling detection. In our experiments, we evaluate three models: a BOW baseline, a logistic supervised classifiers based on word embeddings, and a bidirectional long short-term memory recurrent neural network (LSTM). Our results show that the LSTM model outperforms the other two models, where explicit features with proven effectiveness for similar tasks were encoded.

Lexical Simplification is the task of reducing the lexical complexity of textual documents by replacing difficult words with easier to read (or understand) expressions while preserving the original meaning. The development of robust pipelined multilingual architectures able to adapt to new languages is of paramount importance in lexical simplification. This paper describes and evaluates a modular hybrid linguistic-statistical Lexical Simplifier that deals with the four major Ibero-Romance Languages: Spanish, Portuguese, Catalan, and Galician. The architecture of the system is the same for the four languages addressed, only the language resources used during simplification are language specific.

Emojis are ideograms which are naturally combined with plain text to visually complement or condense the meaning of a message. Despite being widely used in social media, their underlying semantics have received little attention from a Natural Language Processing standpoint. In this paper, we investigate the relation between words and emojis, studying the novel task of predicting which emojis are evoked by text-based tweet messages. We train several models based on Long Short-Term Memory networks (LSTMs) in this task. Our experimental results show that our neural model outperforms a baseline as well as humans solving the same task, suggesting that computational models are able to better capture the underlying semantics of emojis.

In the current context of scientific information overload, text mining tools are of paramount importance for researchers who have to read scientific papers and assess their value. Current citation networks, which link papers by citation relationships (reference and citing paper), are useful to quantitatively understand the value of a piece of scientific work, however they are limited in that they do not provide information about what specific part of the reference paper the citing paper is referring to. This qualitative information is very important, for example, in the context of current community-based scientific summarization activities. In this paper, and relying on an annotated dataset of co-citation sentences, we carry out a number of experiments aimed at, given a citation sentence, automatically identify a part of a reference paper being cited. Additionally our algorithm predicts the specific reason why such reference sentence has been cited out of five possible reasons.

Scientific literature records the research process with a standardized structure and provides the clues to track the progress in a scientific field. Understanding its internal structure and content is of paramount importance for natural language processing (NLP) technologies. To meet this requirement, we have developed a multi-layered annotated corpus of scientific papers in the domain of Computer Graphics. Sentences are annotated with respect to their role in the argumentative structure of the discourse. The purpose of each citation is specified. Special features of the scientific discourse such as advantages and disadvantages are identified. In addition, a grade is allocated to each sentence according to its relevance for being included in a summary. To the best of our knowledge, this complex, multi-layered collection of annotations and metadata characterizing a set of research papers had never been grouped together before in one corpus and therefore constitutes a newer, richer resource with respect to those currently available in the field.

In this paper we present a gold standard dataset for Entity Linking (EL) in the Music Domain. It contains thousands of musical named entities such as Artist, Song or Record Label, which have been automatically annotated on a set of artist biographies coming from the Music website and social network Last.fm. The annotation process relies on the analysis of the hyperlinks present in the source texts and in a voting-based algorithm for EL, which considers, for each entity mention in text, the degree of agreement across three state-of-the-art EL systems. Manual evaluation shows that EL Precision is at least 94%, and due to its tunable nature, it is possible to derive annotations favouring higher Precision or Recall, at will. We make available the annotated dataset along with evaluation data and the code.

Emojis allow us to describe objects, situations and even feelings with small images, providing a visual and quick way to communicate. In this paper, we analyse emojis used in Twitter with distributional semantic models. We retrieve 10 millions tweets posted by USA users, and we build several skip gram word embedding models by mapping in the same vectorial space both words and emojis. We test our models with semantic similarity experiments, comparing the output of our models with human assessment. We also carry out an exhaustive qualitative evaluation, showing interesting results.

WordNet is probably the best known lexical resource in Natural Language Processing. While it is widely regarded as a high quality repository of concepts and semantic relations, updating and extending it manually is costly. One important type of relation which could potentially add enormous value to WordNet is the inclusion of collocational information, which is paramount in tasks such as Machine Translation, Natural Language Generation and Second Language Learning. In this paper, we present ColWordNet (CWN), an extended WordNet version with fine-grained collocational information, automatically introduced thanks to a method exploiting linear relations between analogous sense-level embeddings spaces. We perform both intrinsic and extrinsic evaluations, and release CWN for the use and scrutiny of the community.

During the last decade the amount of scientific information available on-line increased at an unprecedented rate. As a consequence, nowadays researchers are overwhelmed by an enormous and continuously growing number of articles to consider when they perform research activities like the exploration of advances in specific topics, peer reviewing, writing and evaluation of proposals. Natural Language Processing Technology represents a key enabling factor in providing scientists with intelligent patterns to access to scientific information. Extracting information from scientific papers, for example, can contribute to the development of rich scientific knowledge bases which can be leveraged to support intelligent knowledge access and question answering. Summarization techniques can reduce the size of long papers to their essential content or automatically generate state-of-the-art-reviews. Paraphrase or textual entailment techniques can contribute to the identification of relations across different scientific textual sources. This tutorial provides an overview of the most relevant tasks related to the processing of scientific documents, including but not limited to the in-depth analysis of the structure of the scientific articles, their semantic interpretation, content extraction and summarization.

Automatic text summarization, the reduction of a text to its essential content is fundamental for an on-line information society. Although many summarization algorithms exist, there are few tools or infrastructures providing capabilities for developing summarization applications. This paper presents a new version of SUMMA, a text summarization toolkit for the development of adaptive summarization applications. SUMMA includes algorithms for computation of various sentence relevance features and functionality for single and multidocument summarization in various languages. It also offers methods for content-based evaluation of summaries.

Irony, a creative use of language, has received scarce attention from the computational linguistics research point of view. We propose an automatic system capable of detecting irony with good accuracy in the social network Twitter. Twitter allows users to post short messages (140 characters) which usually do not follow the expected rules of the grammar, users tend to truncate words and use particular punctuation. For these reason automatic detection of Irony in Twitter is not trivial and requires specific linguistic tools. We propose in this paper a new set of experiments to assess the relevance of the features included in our model. Our model does not include words or sequences of words as features, aiming to detect inner characteristic of Irony.

Information in newspapers is often showed in the form of numerical expressions which present comprehension problems for many people, including people with disabilities, illiteracy or lack of access to advanced technology. The purpose of this paper is to motivate, describe, and demonstrate a rule-based lexical component that simplifies numerical expressions in Spanish texts. We propose an approach that makes news articles more accessible to certain readers by rewriting difficult numerical expressions in a simpler way. We will showcase the numerical simplification system with a live demo based on the execution of our components over different texts, and which will consider both successful and unsuccessful simplification cases.

Text summarization and information extraction systems require adaptation to new domains and languages. This adaptation usually depends on the availability of language resources such as corpora. In this paper we present a comparable corpus in Spanish and English for the study of cross-lingual information extraction and summarization: the CONCISUS Corpus. It is a rich human-annotated dataset composed of comparable event summaries in Spanish and English covering four different domains: aviation accidents, rail accidents, earthquakes, and terrorist attacks. In addition to the monolingual summaries in English and Spanish, we provide automatic translations and ``comparable'' full event reports of the events. The human annotations are concepts marked in the textual sources representing the key event information associated to the event type. The dataset has also been annotated using text processing pipelines. It is being made freely available to the research community for research purposes.

In this paper we describe the development of a text simplification system for Spanish. Text simplification is the adaptation of a text to the special needs of certain groups of readers, such as language learners, people with cognitive difficulties and elderly people, among others. There is a clear need for simplified texts, but manual production and adaptation of existing texts is labour intensive and costly. Automatic simplification is a field which attracts growing attention in Natural Language Processing, but, to the best of our knowledge, there are no simplification tools for Spanish. We present a prototype for automatic simplification, which shows that the most important structural simplification operations can be successfully treated with an approach based on rules which can potentially be improved by statistical methods. For the development of this prototype we carried out a corpus study which aims at identifying the operations a text simplification system needs to carry out in order to produce an output similar to what human editors produce when they simplify texts.

We present a set of tools and resources for the analysis of interviews during psychotherapy sessions. One of the main components of the work is a dictionary-based text interpretation tool for the Spanish language. The tool is designed to identify a subset of Freudian drives in patient and therapist discourse.

We describe a set of tools, resources, and experiments for opinion classification in business-related datasources in two languages. In particular we concentrate on SentiWordNet text interpretation to produce word, sentence, and text-based sentiment features for opinion classification. We achieve good results in experiments using supervised learning machine over syntactic and sentiment-based features. We also show preliminary experiments where the use of summaries before opinion classification provides competitive advantage over the use of full documents.

Nous étudions différentes méthodes d’évaluation de résumé de documents basées sur le contenu. Nous nous intéressons en particulier à la corrélation entre les mesures d’évaluation avec et sans référence humaine. Nous avons développé FRESA, un nouveau système d’évaluation fondé sur le contenu qui calcule les divergences entre les distributions de probabilité. Nous appliquons notre système de comparaison aux diverses mesures d’évaluation bien connues en résumé de texte telles que la Couverture, Responsiveness, Pyramids et Rouge en étudiant leurs associations dans les tâches du résumé multi-document générique (francais/anglais), focalisé (anglais) et résumé mono-document générique (français/espagnol).

In the context of ontology-based information extraction, identity resolution is the process of deciding whether an instance extracted from text refers to a known entity in the target domain (e.g. the ontology). We present an ontology-based framework for identity resolution which can be customized to different application domains and extraction tasks. Rules for identify resolution, which compute similarities between target and source entities based on class information and instance properties and values, can be defined for each class in the ontology. We present a case study of the application of the framework to the problem of multi-source job vacancy extraction

We describe a number of experiments carried out to address the problem of creating summaries from multiple sources in multiple languages. A centroid-based sentence extraction system has been developed which decides the content of the summary using texts in different languages and uses sentences from English sources alone to create the final output. We describe the evaluation of the system in the recent Multilingual Summarization Evaluation MSE 2005 using the pyramids and ROUGE methods.

We describe the Cubreporter information access system which allows access to news archives through the use of natural language technology. The system includes advanced text search, question answering, summarization, and entity profiling capabilities. It has been designed taking into account the characteristics of the background gathering task.