text stringlengths 4 222k | label int64 0 4 |
|---|---|
Self-supervised learning has emerged as an important training paradigm for learning model parameters which are more generalizable and yield better representations for many down-stream tasks. This typically involves learning through labels that come * Correspondence: tbansal@cs.umass.edu naturally with data, for example... | 0 |
Knowledge about processes is essential for AI systems in order to understand and reason about the world. At the simplest level, even knowing which class of entities play key roles can be useful for tasks involving recognition and reasoning about processes. For instance, given a description "a puddle drying in the sun",... | 0 |
Enormous data in social media has drawn much attention in medical applications. With the rapid development of health language processing, effective systems in mining health information from social media were built to assist pharmacy, diagnosis, nursing, and so on (Paul et al., 2016) (Yang et al., 2012) (Zhou et al., 20... | 0 |
Modeling a composite dialogue (Peng et al., 2017) , which consists of several inherent subtasks, is in high demand due to the complexity of human conversation. For instance, a composite dialogue of making a hotel reservation involves several sub-tasks, such as looking for a hotel that meets the user's constraints, book... | 0 |
During the evaluation campaign for the 2011 International Workshop on Spoken Language Translation (IWSLT-2011) our experimental efforts centered on 1) speech recognition for lecture-like data, 2) improved cross-domain translation using MAP adaptation using corpus distance measures in addition to count-based smoothing, ... | 0 |
In named-entity recognition, the aim is to annotate all occurrences of explicit names, like John (person) and General Motors (organization) in a text, using some defined set of name tags. Machine learning algorithms can be trained to perform this task. However, names manifest themselves quite differently in different d... | 0 |
The Arabic language today is characterized by a complex state of polyglossia. Modern Standard Arabic (MSA) is the official variety of Arabic used primarily in written literal contexts. There is also a large number of dialects whose dominant features are noticeable to Arab-speaking people. The Arabic dialects differ fro... | 0 |
As with many other statistical natural language processing tasks, statistical machine translation (Brown et al., 1993 ) produces high quality results when ample training data is available. This is problematic for so called "low density" language pairs which do not have very large parallel corpora. For example, when wor... | 0 |
With the advent of Web 2.0, individual users have been able to actively participate in the generation of online content via community forums or social media. Online publishing is no longer the realm of large software companies and media organisations, with the Web open and accessible to an ever-larger percentage of the... | 0 |
Comparable corpora contain texts written in different languages that, roughly speaking, "talk about the same thing". In comparison to parallel corpora, ie corpora which are mutual translations, comparable corpora have not received much attention from the research community, and very few methods have been proposed to ex... | 0 |
Analytic verb forms (henceforth AVFs) consist of one or more auxiliaries and a content verb. The auxiliaries can be seen either as marking the content verb with morphological categories or as being part of a multi-word expression, to which the categories are assigned. This is the perspective taken by all standard gramm... | 0 |
Topic models such as latent Dirichlet allocation (LDA) (Blei et al., 2003) are hierarchical probabilistic models of document collections. They can effectively uncover the main themes of corpora by using latent topics learnt from observed collections (Blei, 2012) , however, they neglect semantic information of words. In... | 0 |
Sarcasm is a form of verbal irony that is intended to express contempt or ridicule. Linguistic studies show that the notion of context incongruity is at the heart of sarcasm (Ivanko and Pexman, 2003) . A popular trend in automatic sarcasm detection is semi-supervised extraction of patterns that capture the underlying c... | 0 |
Sense tagging is the task of assigning senses chosen from a computational lexicon to words in context. This is a task where both machines and humans find it difficult to reach an agreement. The problem depends on a variety of factors, ranging from the inherent subjectivity of the task to the granularity of sense discre... | 0 |
Morphological analysis is among the first steps in the natural processing pipeline of morphologically rich languages. Analysis is often preceded by the relatively simpler steps of sentence segmentation and tokenization. Morphological analysis of a token usually yields more than one analysis, in which case a disambiguat... | 0 |
As COVID-19 spreads rapidly around the world, governments have implemented different NPIs to contain the spread of the virus. While effective at slowing down the spread of COVID-19 (Haug et al., 2020) , NPIs such as school and non-essential businesses closures, telecommuting, mask requirements and physical distancing m... | 0 |
Shopping in physical stores is a popular option for many people. Each week, a lot of people enter supermarkets in which they are immersed with many different product choices. In many shopping centers, customer service representatives (CSRs) are employed to answer questions from customers about products. However, a cust... | 0 |
Quantifiers (words like 'some', 'most', 'all') have long been the holy grail of formal semanticists (see Peters et al. (2006) for an overview). More recently, they have caught the attention of cognitive scientists, who showed that these expressions are handled by children quite early in life (Halberda et al., 2008) , e... | 0 |
It is now uncontroversial that dialogue participants construe meaning from utterances on at least as finegrained a level as word-by-word (see Brennan, 2000; Schlesewsky and Bornkessel, 2004, inter alia) . It has also become clear that using more fine-grained incremental processing allows more likeable and interactive s... | 0 |
Many important natural language inferences can be viewed as problems of resolving phonetic, syntactic, semantics or pragmatics ambiguities, based on properties of the surrounding context. It is generally accepted that a learning component must have a central role in resolving these context sensitive ambiguities, and a ... | 0 |
Universal Decompositional Semantics (UDS) (White et al., 2016) is a contemporary semantic representation of text (Abend and Rappoport, 2017) that forgoes traditional inventories of semantic categories in favor of bundles of simple, interpretable properties. In particular, UDS includes a practical implementation of Dowt... | 0 |
English is a configurational language, so grammatical functions are mostly expressed through word order and function words, rather than with inflectional morphology. Most English verbs have four forms, and none have more than five. Most of the world's languages have far richer inflectional morphology, some with million... | 0 |
FrameNet (FN) (Baker et al., 1998 ) is a lexicalsemantic resource manually built by FN experts. It embodies the theory of frame semantics (Fillmore, 1976) : the frames capture units of meaning corresponding to prototypical situations. Besides FN's definitions of frame-specific roles and frame-evoking elements that are ... | 0 |
In recent years, the explosion of social media has changed the relation between the users and the web. The world has become closer and more "realtime" than ever. People have increasingly been part of virtual society where they have created their content, shared it, interacted with others in different ways and at a very... | 0 |
Arabic is the fourth most widely spoken language in the world (Nwesri et al., 2005) . It is a morphologically and syntactically rich language. Arabic morphological and syntactic analyses have gained the focus of Arabic natural language processing research for a long time in order to achieve the automated understanding ... | 0 |
While speech recognition is an easy task for humans even under difficult acoustic conditions, current ASR systems still cannot compete with humans (Potamianos et al., 2003) . This is especially true in human-robot interaction, where one has to deal with spontaneous speech effects, noisy environments, communicative gest... | 0 |
An ontology is a knowledge base with information about concepts existing in the world or domain, their properties, and how they relate to each other. Three principal reasons to use an ontology in machine translation (MT) are to enable source language analyzers and target language generators to share knowledge, to store... | 0 |
Discourse Representation Theory (DRT) is a popular theory of meaning representation (Kamp, 1981; Kamp and Reyle, 2013; Asher, 1993; Asher et al., 2003) designed to account for a variety of linguistic phenomena within and across sentences. The basic meaning-carrying units in DRT are Discourse Representation Structures (... | 0 |
A recent success story of NLP, BERT (Devlin et al., 2018) stands at the crossroad of two key innovations that have brought about significant improvements over previous state-of-the-art results. On the one hand, BERT models are an instance of contextual embeddings (McCann et al., 2017; Peters et al., 2018) , which have ... | 0 |
Since their introduction as a method for aligning inputs and outputs in neural machine translation, attention mechanisms (Bahdanau et al., 2014) have emerged as effective components in various neural network architectures. Attention works by aggregating a set of tokens via a weighted sum, where the attention weights ar... | 0 |
Widely adopted Transformer architecture (Vaswani et al., 2017) has obviated the need for sequential processing of the input that is enforced in traditional Recurrent Neural Networks (RNN). As a result, compared to a single-layered LSTM or RNN model, a single-layered Transformer model is computationally more efficient, ... | 0 |
Empathetic conversation studies have been coming to the forefront in recent years owing to the increasing interest in dialogue systems. Empathetic dialogues not only provide dialogue partners with highly relevant contents but also project their feelings and convey a special emotion, that is, empathy. As revealed by pre... | 0 |
The CKY algorithm is an O |G|n 3 dynamic programming algorithm for finding all of the possible derivations of a sentence in a context-free language. Its complexity depends on both the sentence length n and the size of the grammar |G|. Methods for improving parsing accuracy typically increase the size of the grammar (Kl... | 0 |
Transformer-based models have shown great power in various natural language processing (NLP) tasks. Pre-trained with gigabytes of unsupervised data, these models usually have hundreds of millions of parameters. For instance, the BERT-base model has 109M parameters, with the model size of 400+MB if represented in 32-bit... | 0 |
Most resources for Arabic have been developed for Modern Standard Arabic (MSA) since it is the official language in most Arabic speaking countries. MSA is used in media coverage, politics, books, and even online. However, in each individual Arabic country, the predominant language used for everyday conversation (in rea... | 0 |
Current probabilistic parsers rely on treebanks in order to estimate their parameters. Such data is known to be difficult and expensive to produce. One can also observe that the quality of parsers increase when they model complex lexicosyntactic phenomena. Unfortunately, increasing the precision of models is quickly co... | 0 |
The locally bound reflexives have been explained by the traditional binding theory in Chomsky (1981) , R&R's predicate based theory (MR 1993) , and the recent derivational theory (Hornstein 2001) . Questions are raised with regard to the reflexive forms that seem to be bound across the clause boundary. Long-distance re... | 0 |
While it used to take decades for machine learning models to surpass estimates of human performance on benchmark tasks, that milestone is now routinely reached within just a few years for newer datasets (see Figure 1) . As with the rest of AI, NLP has advanced rapidly thanks to improvements in computational power, as w... | 0 |
A standard metric in automatic multi-and single-document summarisation is ROUGE. Certain variants of the metric have been shown to correlate well with human judgments (Lin, 2004) . In particular, ROUGE-2 is the ROUGE variant shown to correlate best with human judgments.While it is impossible to directly optimise for th... | 0 |
In the era where massive texts are produced every day, machines are indispensable to assist with daily life. However, to be truly helpful, machines must understand the meaning of texts (natural language understanding: NLU). An essential task for NLU is Recognizing Textual Entailment (RTE), which is also known as Natura... | 0 |
Sentiment classification is an important task of natural language processing (NLP), aiming to classify the sentiment polarity of a given text as positive, negative, or more fine-grained classes. It has obtained considerable attention due to its broad applications in natural language processing (Hao et al., 2012; . Most... | 0 |
Slot-filling is a key component of task-driven dialog systems, providing a way for systems to extract key properties from user queries. For example, a slot-filling model can extract the tokens "New York" as a TO LOCATION slot in the query "book a flight to New York". Standard slot-filling models train or finetune on la... | 0 |
Discourse planning is a subtask of Natural Language Generation (NLG), concerned with determining the ordering of messages in a document and the discourse relations that hold among them (Reiter and Dale, 2000) . Early approaches to discourse planning used manually written rules, often based on schemas (McKeown, 1985) or... | 0 |
MedSLT is a medical speech translation system. It allows a doctor to ask diagnosis questions in medical subdomains, such as headaches, abdominal pain, etc, covering a wide range of questions that doctors generally ask their patients. The grammarbased architecture, built using specialization from reusable general gramma... | 0 |
Text classification is one of the most fundamental tasks in Natural Language Processing, and has found its way into a wide spectrum of NLP applications, ranging from email spam detection and social media analytics to sentiment analysis and data mining. Over the past couple of decades, supervised statistical learning me... | 0 |
Multiword expressions (MWEs), commonly referred to as collocations, 1 are idiosyncratic sequences of words whose idiosyncrasy can be broadly classified into semantic, statistical, and syntactic classes. Semantic idiosyncrasy (also referred to as non-compositionality) means that the meaning of an MWE cannot be inferred ... | 0 |
This paper describes LIMSI's submissions to the shared translation task of the Eighth Workshop on Statistical Machine Translation. LIMSI participated in the French-English, German-English and Spanish-English tasks in both directions. For this evaluation, we used n-code, an open source inhouse Statistical Machine Transl... | 0 |
Existing social media analysis systems are hampered by their inability to accurately detect and interpret figurative language. This is particularly relevant in domains like the social sciences and politics, in which the use of figurative communication devices such as verbal irony (roughly, sarcasm) is common. Sarcasm i... | 0 |
Several general-purpose rule-based generation systems have been developed, some of which are available publicly (cf. Elhadad, 1992) . Unfortunately these systems, because of their generality, can be difficult to adapt to small, task-oriented applications. Bateman and Henschel (1999) have described a lower cost and more... | 0 |
In the paper we propose a method to integrate the logical analysis of sentences with the linguistic approach to semantics, exploiting the complex valency frames (CVFs) in the VerbaLex verb valency lexicon, see (Hlaváčková, Horák, Kadlec 2006) . To this end we first present a brief survey of the logic we are going to us... | 0 |
Computer-assisted translation (CAT) is an important frontier where current MT technology meets professional translators. While MT is generally not yet able to provide output that is suitable for publication without human intervention, CAT is an ideal scenario, because human feedback is always available. In fact, while ... | 0 |
Topic modeling has become a popular tool for applied research such as social media analysis, as it facilitates the exploration of large documentcollections and yields insights that would not be accessible by manual methods (Sinnenberg et al., 2017; Karami et al., 2020) . However, social media data can be challenging to... | 0 |
Counterfactual refers to a conditional statement in which the first clause is a past tense subjunctive statement expressing something contrary to fact, as in "If I had studied harder, I might have passed the exam". Sometimes it can be used in practice to support the assumption that if events occurred differently in the... | 0 |
Continuous vector representations of words, also known as word embeddings, have been used as features for all kinds of NLP tasks such as Information Extraction (Lample et al., 2016; Zeng et al., 2014; , Semantic Parsing (Chen and Manning, 2014; Zhou and Xu, 2015; Konstas et al., 2017) , Sentiment Analysis (Socher et al... | 0 |
The articulation of mathematical arguments is a fundamental part of scientific reasoning and communication. Across many scientific disciplines, expressing relations and inter-dependencies between quantities is at the centre of its argumentation. One of the particular linguistic elements used for such argumentation is v... | 0 |
In this paper, we discuss a developing approach towards modeling peer-to-peer communication using multiple modalities, e.g., language, gesture, vision, and action. This platform integrates a multimodal model of semantics (Multimodal Semantic Simulations, MSS) (Pustejovsky and Krishnaswamy, 2016; Krishnaswamy et al., 20... | 0 |
Speech translation is an important field that becomes more relevant with every improvement to its component technologies of automatic speech recognition (ASR) and machine translation (MT). It enables exciting applications like live machine interpretation (Cho and Esipova, 2016; and automatic foreign-language subtitling... | 0 |
The success of corpus-based approaches to discourse ultimately depends on whether one is able to acquire a large volume of data annotated for discourse-level information. However, to acquire merely a few hundred texts annotated for discourse information is often impossible due to the enormity of the haman labor require... | 0 |
Dans l'idée de disposer de corpus annotés pour le français, nous avons l'objectif de développer le FDTB (French Discourse Tree Bank), un corpus annoté pour l'analyse discursive. Le FDTB s'inspire du PDTB (Penn Discourse Tree Bank, (PDTB Group, 2008)) qui ajoute une couche d'annotation discursive (manuelle) sur le PTB-v... | 0 |
This paper investigates the use of indexical determiners (i.e. determiners employed for direct references to objects and that include a pointing gesture) by Dutch, Portuguese and English speakers. A comparison of the use of Dutch and English demonstratives in terms of the accessibility of the target by Piwek and Cremer... | 0 |
Mi'kmaq is an Indigenous language spoken primarily in Eastern Canada (Johnson, 1996) . It is polysynthetic and verb-oriented, and in the Eastern Algonquian language family. Mi'kmaq has roughly 8,000 speakers in Canada, 1 and is a low-resource language. There are Mi'kmaq dictionaries (Rand, 1888; DeBlois, 1996) and tran... | 0 |
Recent years have witnessed a trend moving from sentence-level neural machine translation (Sen-NMT) to its document-level counterpart (Doc-NMT). SenNMT inevitably suffers from translation errors related with document phenomena (Maruf et al., 2021) and delivers obviously inferior performance when compared against human ... | 0 |
Claims present in daily news are unfiltered and potentially of great value, but can also have negative effects when misinformation is widespread. The COVID-19 pandemic is a crucial example of when false claims can be particularly harmful, with the torrent of misinformation impacting public perception. For example, a cl... | 0 |
Discourse connectives are closed-class lexical items that are known to provide very useful information for tasks like discourse parsing in the style of RST (e.g., (Hernault et al., 2010) ) or PDTB (e.g., (Lin et al., 2014) ), relation extraction (e.g., finding causal statements in biomedical text, (van der Horn et al.,... | 0 |
Neural machine translation (NMT) systems (Sutskever et al., 2014; Bahdanau et al., 2015; Vaswani et al., 2017) have demonstrated superior performance with large amounts of parallel data. However, the performance of most existing NMT systems will degrade when the labeled data is limited (Koehn and Knowles, 2017; . To ad... | 0 |
In this paper, we will continue investigation into Denial of Expectation (DofE) across turns in dialogue when signalled by "but", following work by Thomas and Matheson (2003) , and claim that these denied expectations need not be causal only. That is, we investigate two hypotheses: (1), that "but" can deny noncausal re... | 0 |
With the advancement in Artificial Intelligence (AI), dialogue systems have become a prominent part in today's virtual assistant, which helps users to converse naturally with the system for effective task completion. Dialogue systems focus on two broad categories -open domain conversations with casual chit chat and goa... | 0 |
Quantifying the uncertainty of machine learning models is an important aspect of trustworthy, reliable, and accountable natural language understanding (NLU) systems. Obtaining measures of uncertainty in predictions (also known as uncertainty estimations, UE) helps to detect out-of-domain (Malinin and Gales, 2018), adve... | 0 |
Lexical simplification (LS) is the task of replacing difficult words with simple words in a text, while preserving its meaning and grammaticality. It aims to produce output text that is easier to understand for readers with special needs, such as language learners, children (Kajiwara et al., 2013) , and those with lang... | 0 |
The capitalization task, also known as truecasing (Lita et al., 2003) , consists of rewriting each word of an input text with its proper case information. The capitalization of a word sometimes depends on its current context, and the intelligibility of texts is strongly influenced by this information. Different practic... | 0 |
Metaphors play a special role in human language and thought, as they evoke a complex array of hidden connotations, past experiences, feelings, and humor, in the service of helping the speaker convey their message in a way that is easier to relate to. However, by their very nature, metaphors continue to pose a challenge... | 0 |
The segmentation of Japanese words is one of the main challenges in the automatic processing of Japanese text. Unlike English text which has spaces that separate consecutive words, there are no such word boundary indicators in sentences of Japanese (kanji and kana) text.The algorithms used to obtain robust segmentation... | 0 |
In the framework of the CoNLL08 shared task (Surdeanu et al., 2008) , a system takes POS tagged sentences as input and produces sentences parsed for syntactic and semantic dependencies as output. A syntactic dependency is represented by an ID of head word and a dependency relation between the head word and its modifier... | 0 |
Finding simple and non-recursive base Noun Phrase (baseNP) is an important subtask for many natural language processing applications, such as partial parsing, information retrieval and machine translation. A baseNP is a simple noun phrase that does not contain other noun phrase recursively, for example, the elements wi... | 0 |
Recent studies demonstrate that deep neural networks are vulnerable to adversarial examples, intentionally crafted to fool the models. Although generating adversarial examples for texts has shown to be more challenging than for images due to their discrete nature, many methods have been proposed to generate adversarial... | 0 |
Learning to communicate is a key capacity of intelligent agents. Research on enabling a machine to have meaningful and natural conversations with humans plays a fundamental role in developing artificial general intelligence, as can be seen in the formulation of Turing test (Turing, 1950) . Recently open-domain or non-t... | 0 |
This paper describes a hybrid Machine Translation (MT) system built for translating from English to German in the domain of technical documentation. The system builds upon the general architecture described in , but in the current version several components have been improved or replaced. As detailed in the previous pa... | 0 |
The staggering amount of machine readable text available on today's Internet calls for increasingly powerful text and language processing methods. This need has fuelled the search for more subtle and sophisticated representations of language meaning, and methods for learning such models. Two well-researched but prima-f... | 0 |
This report presents the participation of Tintin (Centre for English Corpus Linguistics) in Task 4 of SemEval 2019 entitled Hyperpartisan News Detection. This task is defined as follows by the organizers 1 : "Given a news article text, decide whether it follows a hyperpartisan argumentation, i.e., whether it exhibits b... | 0 |
One of the most interesting topics in language studies has been on how speakers' gender and age differences influence their communicative behaviour. Transcriptions of real-world, naturallyoccurring conversations provide us a window to examine such differences in talk-in-interaction.Gendered and age-salient elements of ... | 0 |
Code-switching or code-mixing is a sociolinguistic phenomenon, where multilingual speakers switch back and forth between two or more common languages or language varieties in a single utterance 1 . The phenomenon is mostly prevalent in spoken language and in informal settings on social media such as in news groups, blo... | 0 |
Pre-trained language models have received great interest in the natural language processing (NLP) community in the last recent years (Dai and Le, 2015; Radford, 2018; Howard and Ruder, 2018; Baevski et al., 2019; . These models are trained in a semi-supervised fashion to learn a general language model, for example, by ... | 0 |
Talmy (2000 a&b) introduced the idea of complex events that he termed 'macro-events'. He said that such events can be conceptualized as being comprised of two simpler events and the relation between them. For example, (1a), shown below, can be conceptualized as one event, and thus expressed in one sentence or phrase, b... | 0 |
Corpus annotation is a time-consuming and a costly activity, thus impeding the development of large language resources. In order to favour the spread and the reuse of such valuable data, a crucial recommendation is to follow established annotation standards. Nonetheless, many areas of Natural Language Processing are st... | 0 |
Understanding events and their participants is a core NLP task, and SRL is the standard approach for identification and labeling of these events in text. SRL systems (Täckström et al., 2015; Roth and Woodsend, 2014) have benefited NLP applications, and many approaches have been proposed to transfer semantic roles from ... | 0 |
Currently access to institutional repositories is gained using dedicated web interfaces where users can enter keywords. In many cases this approach is rather cumbersome for users who are required to learn a syntax specific to that particular interface. A solution to this problem is offered by question answering (QA), a... | 0 |
Les listes de mots annotés sont l'un des outils fréquemment utilisés dans l'analyse des sentiments pour détecter une humeur ou classer des textes en fonction des émotions qui y sont exprimées. Ces listes contiennent pour chaque mot son évaluation émotionnelle qui peut être représentée par un ensemble de scores numériqu... | 0 |
A knowledgebase which systemizes lexical and conceptual information of human knowledge is a basic infrastructure for Natural Language Processing (NLP) applications. Wordnets, pioneered by the Princeton WordNet (WN, Fellbaum 1998) , and greatly enriched by EuroWordnet (EWN, Vossen 1998) , have become the standard for a ... | 0 |
Semantic role labeling (SRL) is one of the fundamental tasks of natural language processing. In a sense, it continues from where syntactic parsing ends: it identifies the events and participants, such as agents and patients, present in a sentence, and therefore it is an essential step in automatically processing the se... | 0 |
In the Japanese-Chinese machine translation task, reordering is the main problem due to substantial differences in sentence structures between these two languages. For example, Japanese has a subject-object-verb (SOV) structure, while, Chinese has a subject-verb-object (SVO) structure.The pre-ordering technology is one... | 0 |
Interlingua based MT (Nirenburg, 1994; Dorr et al., 2010) relies on the principle that every language in the world can be mapped to a common linguistic representation. Further, given this representation, it should be possible to decode a target sentence in any language. This implies that given n languages we just need ... | 0 |
Research in Machine Translation (MT) has seen significant advances in recent years thanks to improvements in modeling, and in particular neural models (Sutskever et al., 2014; Bahdanau et al., 2015; Gehring et al., 2016; Vaswani et al., 2017) , as well as the availability of large parallel corpora for training (Tiedema... | 0 |
Event trigger labeling is the task of identifying the main word tokens that express mentions of prespecified event types in running text. For example, in "20 people were wounded in Tuesday's airport blast", "wounded" is a trigger of an Injure event and "blast" is a trigger of an Attack. The task both detects trigger to... | 0 |
Dependency structure represents the grammatical relations that hold between the words in a sentence. It encodes semantic relations directly, and has the best inter-lingual phrasal cohesion properties (Fox, 2002) . Those attractive characteristics make it pos-sible to improve translation quality by using dependency stru... | 0 |
TimeML 1 is a markup language which allows us, among other things, to annotate and analyse the temporal structure of a text and to represent it diagrammatically, e.g. in the T-Box format (Verhagen, 2007) . It has been used to annotate TimeBank (Pustejovsky et al., 2003) , 2 a collection of news articles from various so... | 0 |
Our production and comprehension of language is a multi-layered computational process. Humans carry out high-level semantic tasks effortlessly by subconsciously using a vast inventory of complex linguistic devices, while simultaneously integrating their background knowledge, to reason about reality. An ideal computatio... | 0 |
Online communications are increasingly becoming fast-paced and frequent, and hidden in these abundant user-generated social media posts are insights for understanding users and their preferences. However, these social media posts often come in unstructured text or images, making massive-scale opinion mining extremely c... | 0 |
The Evolex project 1 brings together researchers in Psycholinguistics and Natural Language Processing (NLP) and focuses on lexical access and lexical relations by pursuing a threefold objective:(1) to propose a new computerised tool for assessing lexical access in population with or without language deficits; (2) to co... | 0 |
Human computer conversation has been an important and challenging task in NLP and AI since the Turing Test was proposed in 1950 (Turing, 1950) . Recently, with the rapid growth of social conversation data available on the Internet, data-driven chatbots are able to learn to generate responses directly and have attracted... | 0 |
Neural Machine Translation (NMT) performs poorly without large training corpora (Koehn and Knowles, 2017) . Domain adaptation is required when there is sufficient data in the desired language pair but insufficient data in the desired domain (the topic, genre, style or level of formality). This work focuses on the super... | 0 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.