_id stringlengths 4 10 | text stringlengths 0 18.4k | title stringlengths 0 8.56k |
|---|---|---|
d14731867 | Today, the named entity recognition task is considered as fundamental, but it involves some specific difficulties in terms of annotation. Those issues led us to ask the fundamental question of what the annotators should annotate and, even more important, for which purpose. We thus identify the applications using named entity recognition and, according to the real needs of those applications, we propose to semantically define the elements to annotate. Finally, we put forward a number of methodological recommendations to ensure a coherent and reliable annotation scheme. | Towards a Methodology for Named Entities Annotation |
d14691484 | This paper introduces a novel method for joint unsupervised aquisition of verb subcategorization frame (SCF) and selectional preference (SP) information. Treating SCF and SP induction as a multi-way co-occurrence problem, we use multi-way tensor factorization to cluster frequent verbs from a large corpus according to their syntactic and semantic behaviour. The method extends previous tensor factorization approaches by predicting whether a syntactic argument is likely to occur with a verb lemma (SCF) as well as which lexical items are likely to occur in the argument slot (SP), and integrates a variety of lexical and syntactic features, including co-occurrence information on grammatical relations not explicitly represented in the SCFs. The SCF lexicon that emerges from the clusters achieves an F-score of 68.7 against a gold standard, while the SP model achieves an accuracy of 77.8 in a novel evaluation that considers all of a verb's arguments simultaneously.TITLE AND ABSTRACT IN FRENCHFactorisation de tenseurs à plusieurs dimensions pour l'acquisition lexicale non superviséeCet article présente une méthode originale pour l'acquisition simultanée de cadres de souscatégorisation (subcategorization frames) et de restrictions de sélection (selectional preferences) appliquée au lexique verbal. L'induction simultanée de ces deux types d'information est vue comme un problème de cooccurrence à plusieurs dimensions. On introduit donc une méthode de factorisation de tenseurs, afin de classer les verbes fréquents d'un grand corpus suivant leur comportement syntaxique. L'approche est fondée sur un ensemble de traits de nature syntaxique et lexicale, y compris des informations de cooccurrence au sein des relations grammaticales qui ne sont pas explicitement représentées dans les schémas de sous-catégorisation. Le dictionnaire de sous-catégorisation produit par la méthode de classification obtient une F-mesure de 68,7 lors de l'évaluation face à un dictionnaire de référence tandis que les restrictions de sélection ont une exactitude (accuracy) de 77,8 en tenant compte de tous les arguments simultanément. | Multi-way Tensor Factorization for Unsupervised Lexical Acquisition |
d57145 | MotivationThe traditional tri-partition syntax/semantics/pragmatics is commonly used in most of the computer systems that aim at the simulation of the human understanding of Natural Language (NL). This conception does not reflect the flexible and creative manner that humans use in reality to interpret texts. Generally speaking, formal NL semantics is referential i.e. it assumes that it is possible to create a static discourse universe and to equate the objects of this universe to the (static) meanings of words. The meaning of a sentence is then built from the meanings of the words in a compositional process and the semantic interpretation of a sentence is reduced to its logical interpretation based on the truth conditions. The very difficult task of adapting the meaning of a sentence to its context is often left to the pragmatic level, and this task requires to use a huge amount of common sense knowledge about the domain. This approach is seriously challenged (see for example [4][14]). It has been showed that the above tri-partition is very artificial because linguistic as well as extra-linguistic knowledge interact in the same global process to provide the necessary elements for understanding. Linguistic phenomena such as polysemy, plurals, metaphors and shifts in meaning create real difficulties to the referential approach of the NL semantics discussed above. As an alternative solution to these problems,[4]proposes an inferential approach to the NL semantics in which words trigger inferences depending on the context of their apparition. In the same spirit we claim that understanding a NL text is a reasoning process based on our knowledge about the norms 1 of its domain i.e. what we generally expect to happen in normal situations. But what kind of reasoning is needed for natural language semantics?The answer to this question is based on the remark that texts seldom provide normal details that are assumed to be known to the reader. Instead, they focus on abnormal situations or at least on events that cannot be inferred by default from the text by an ordinary reader. A central issue in the human understanding of NL is the ability to infer systematically and easily an amount of implicit information necessary to answer indirect questions about the text. The consequences resulting from truth-based entailments are logically valid but they are poor and quite limited. Those obtained by a norm-based approach are defeasible: they are admitted as long as the text does not mention explicit elements that contradict them. However they provide richer information and enable a deeper understanding of the text. That is why the norm-based reasoning must be non-monotonic. In addition to this central question, the representation language must take into account a number of modalities (including the temporal aspect) that are very useful to answer different questions on NL texts.The next section gives a general logical framework to represent in a first order language the necessary knowledge about a domain and allows non-monotonic reasoning. Section 3 shows how to implement our representation language fragment in the formalism of Answer Set Programming by transforming them into extended logic programs. In section 4, we discuss the use of our language in 1 In A.I, the word norm is commonly used in the « normative » sens. Here, it is rather used in the « normal » sens. | Using Answer Set Programming in an Inference-Based approach to Natural Language Semantics |
d10619801 | Identifying mathematical relations expressed in text is essential to understanding a broad range of natural language text from election reports, to financial news, to sport commentaries to mathematical word problems. This paper focuses on identifying and understanding mathematical relations described within a single sentence. We introduce the problem of Equation Parsing -given a sentence, identify noun phrases which represent variables, and generate the mathematical equation expressing the relation described in the sentence. We introduce the notion of projective equation parsing and provide an efficient algorithm to parse text to projective equations. Our system makes use of a high precision lexicon of mathematical expressions and a pipeline of structured predictors, and generates correct equations in 70% of the cases. In 60% of the time, it also identifies the correct noun phrase → variables mapping, significantly outperforming baselines. We also release a new annotated dataset for task evaluation. | EQUATION PARSING : Mapping Sentences to Grounded Equations |
d8462706 | We introduce the Treebank of Learner English (TLE), the first publicly available syntactic treebank for English as a Second Language (ESL). The TLE provides manually annotated POS tags and Universal Dependency (UD) trees for 5,124 sentences from the Cambridge First Certificate in English (FCE) corpus. The UD annotations are tied to a pre-existing error annotation of the FCE, whereby full syntactic analyses are provided for both the original and error corrected versions of each sentence. Further on, we delineate ESL annotation guidelines that allow for consistent syntactic treatment of ungrammatical English. Finally, we benchmark POS tagging and dependency parsing performance on the TLE dataset and measure the effect of grammatical errors on parsing accuracy. We envision the treebank to support a wide range of linguistic and computational research on second language acquisition as well as automatic processing of ungrammatical language 1 . | Universal Dependencies for Learner English |
d226283468 | In this paper, we study the grounding skills required to answer spatial questions asked by humans while playing the GuessWhat?! game. We propose a classification for spatial questions dividing them into absolute, relational, and group questions. We build a new answerer model based on the LXMERT multimodal transformer and we compare a baseline with and without visual features of the scene. We are interested in studying how the attention mechanisms of LXMERT are used to answer spatial questions since they require putting attention on more than one region simultaneously and spotting the relation holding among them. We show that our proposed model outperforms the baseline by a large extent (9.70% on spatial questions and 6.27% overall). By analyzing LXMERT errors and its attention mechanisms, we find that our classification helps to gain a better understanding of the skills required to answer different spatial questions. | They Are Not All Alike: Answering Different Spatial Questions Requires Different Grounding Strategies |
d252624541 | Subjective factors affect our familiarity with different words. Our education, mother tongue, dialect or social group all contribute to the words we know and understand. When asking people to mark words they understand some words are unanimously agreed to be complex, whereas other annotators universally disagree on the complexity of other words. In this work, we seek to expose this phenomenon and investigate the factors affecting whether a word is likely to be subjective, or not. We investigate two recent word complexity datasets from shared tasks. We demonstrate that subjectivity is present and describable in both datasets. Further we show results of modelling and predicting the subjectivity of the complexity annotations in the most recent dataset, attaining an F1-score of 0.714. | Agree to Disagree: Exploring Subjectivity in Lexical Complexity |
d249240376 | In recent years, the natural language processing (NLP) community has given increased attention to the disparity of efforts directed towards high-resource languages over low-resource ones. Efforts to remedy this delta often begin with translations of existing English datasets into other languages. However, this approach ignores that different language communities have different needs. We consider a group of low-resource languages, Creole languages. Creoles are both largely absent from the NLP literature, and also often ignored by society at large due to stigma, despite these languages having sizable and vibrant communities. We demonstrate, through conversations with Creole experts and surveys of Creole-speaking communities, how the things needed from language technology can change dramatically from one language to another, even when the languages are considered to be very similar to each other, as with Creoles. We discuss the prominent themes arising from these conversations, and ultimately demonstrate that useful language technology cannot be built without involving the relevant community. | What a Creole Wants, What a Creole Needs |
d2390655 | Idiomatic expressions are plentiful in everyday language, yet they remain mysterious, as it is not clear exactly how people learn and understand them. They are of special interest to linguists, psycholinguists, and lexicographers, mainly because of their syntactic and semantic idiosyncrasies as well as their unclear lexical status. Despite a great deal of research on the properties of idioms in the linguistics literature, there is not much agreement on which properties are characteristic of these expressions. Because of their peculiarities, idiomatic expressions have mostly been overlooked by researchers in computational linguistics. In this article, we look into the usefulness of some of the identified linguistic properties of idioms for their automatic recognition. Specifically, we develop statistical measures that each model a specific property of idiomatic expressions by looking at their actual usage patterns in text. We use these statistical measures in a type-based classification task where we automatically separate idiomatic expressions (expressions with a possible idiomatic interpretation) from similar-on-the-surface literal phrases (for which no idiomatic interpretation is possible). In addition, we use some of the measures in a token identification task where we distinguish idiomatic and literal usages of potentially idiomatic expressions in context. | Unsupervised Type and Token Identification of Idiomatic Expressions |
d44077352 | This paper describes the UMDSub system that participated in Task 2 of SemEval-2018. We developed a system that predicts an emoji given the raw text in a English tweet. The system is a Multi-channel Convolutional Neural Network based on subword embeddings for the representation of tweets. This model improves on character or word based methods by about 2%. Our system placed 21st of 48 participating systems in the official evaluation. | UMDSub at SemEval-2018 Task 2: Multilingual Emoji Prediction Multi-channel Convolutional Neural Network on Subword Embedding |
d9918276 | Existing discourse research only focuses on the monolingual languages and the inconsistency between languages limits the power of the discourse theory in multilingual applications such as machine translation. To address this issue, we design and build a bilingual discource corpus in which we are currently defining and annotating the bilingual elementary discourse units (BEDUs). The BEDUs are then organized into hierarchical structures. Using this discourse style, we have annotated nearly 20K LDC sentences. Finally, we design a bilingual discourse based method for machine translation evaluation and show the effectiveness of our bilingual discourse annotations. | A Bilingual Discourse Corpus and Its Applications |
d13679932 | Counterfactual learning from human bandit feedback describes a scenario where user feedback on the quality of outputs of a historic system is logged and used to improve a target system. We show how to apply this learning framework to neural semantic parsing. From a machine learning perspective, the key challenge lies in a proper reweighting of the estimator so as to avoid known degeneracies in counterfactual learning, while still being applicable to stochastic gradient optimization.To conduct experiments with human users, we devise an easy-to-use interface to collect human feedback on semantic parses. Our work is the first to show that semantic parsers can be improved significantly by counterfactual learning from logged human feedback data. | Improving a Neural Semantic Parser by Counterfactual Learning from Human Bandit Feedback |
d5208259 | Treebanks, such as the Penn Treebank (PTB), offer a simple approach to obtaining a broad coverage grammar: one can simply read the grammar off the parse trees in the treebank.While such a grammar is easy to obtain, a square-root rate of growth of the rule set with corpus size suggests that the derived grammar is far from complete and that much more treebanked text would be required to obtain a complete grammar, if one exists at some limit.However, we offer an alternative explanation in terms of the underspecification of structures within the treebank.This hypothesis is explored by applying an algorithm to compact the derived grammar by eliminating redundant rules -rules whose right hand sides can be parsed by other rules.The size of the resulting compacted grammar, which is significantly less than that of the full treebank grammar, is shown to approach a limit.However, such a compacted grammar does not yield very good performance figures.A version of the compaction algorithm taking rule probabilities into account is proposed, which is argued to be more linguistically motivated.Combined with simple thresholding, this method can be used to give a 58% reduction in grammar size without significant change in parsing performance, and can produce a 69% reduction with some gain in recall, but a loss in precision. | Compacting the Penn Treebank Grammar |
d16749512 | We present a discriminative model that directly predicts which set of phrasal translation rules should be extracted from a sentence pair. Our model scores extraction sets: nested collections of all the overlapping phrase pairs consistent with an underlying word alignment. Extraction set models provide two principle advantages over word-factored alignment models. First, we can incorporate features on phrase pairs, in addition to word links. Second, we can optimize for an extraction-based loss function that relates directly to the end task of generating translations. Our model gives improvements in alignment quality relative to state-of-the-art unsupervised and supervised baselines, as well as providing up to a 1.4 improvement in BLEU score in Chinese-to-English translation experiments. | Discriminative Modeling of Extraction Sets for Machine Translation |
d3411445 | Distributed word representations, or word vectors, have recently been applied to many tasks in natural language processing, leading to state-of-the-art performance. A key ingredient to the successful application of these representations is to train them on very large corpora, and use these pre-trained models in downstream tasks. In this paper, we describe how we trained such high quality word representations for 157 languages. We used two sources of data to train these models: the free online encyclopedia Wikipedia and data from the common crawl project. We also introduce three new word analogy datasets to evaluate these word vectors, for French, Hindi and Polish. Finally, we evaluate our pre-trained word vectors on 10 languages for which evaluation datasets exists, showing very strong performance compared to previous models. | Learning Word Vectors for 157 Languages |
d137267 | The multidimensional, heterogeneous, and temporal nature of speech databases raises interesting challenges for representation and query. Recently, annotation graphs have been proposed as a general-purpose representational framework for speech databases. Typical queries on annotation graphs require path expressions similar to those used in semistructured query languages. However, the underlying model is rather different from the customary graph models for semistructured data: the graph is acyclic and unrooted, and both temporal and inclusion relationships are important. We develop a query language and describe optimization techniques for an underlying relational representation. | Towards A Query Language for Annotation Graphs |
d3662564 | In this paper, we introduce DuReader, a new large-scale, open-domain Chinese machine reading comprehension (MRC) dataset, aiming to tackle real-world MRC problems. In comparison to prior datasets, DuReader has the following characteristics: (a) the questions and the documents are all extracted from real application data, and the answers are human generated; (b) it provides rich annotations for question types, especially yes-no and opinion questions, which take a large proportion in real users' questions but have not been well studied before; (c) it provides multiple answers for each question. The first release of DuReader contains 200k questions, 1,000k documents, and 420k answers, which, to the best of our knowledge, is the largest Chinese MRC dataset so far. Experimental results show there exists big gap between the state-of-the-art baseline systems and human performance, which indicates DuReader is a challenging dataset that deserves future study. The dataset and the code of the baseline systems are publicly available now 1 . | DuReader: a Chinese Machine Reading Comprehension Dataset from Real-world Applications |
d12456993 | This work explores normalization for parser adaptation. Traditionally, normalization is used as separate pre-processing step. We show that integrating the normalization model into the parsing algorithm is beneficial. This way, multiple normalization candidates can be leveraged, which improves parsing performance on social media. We test this hypothesis by modifying the Berkeley parser; out-ofthe-box it achieves an F1 score of 66.52. Our integrated approach reaches a significant improvement with an F1 score of 67.36, while using the best normalization sequence results in an F1 score of only 66.94. | Parser Adaptation for Social Media by Integrating Normalization |
d1842 | This paper shows how to induce an N-best translation lexicon from a bilingual text corpus using statistical properties of the corpus together with four external knowledge sources. The knowledge sources are cast as lters, so that any subset of them can be cascaded in a uniform framework. A new objective evaluation measure is used to compare the quality of lexicons induced with di erent lter cascades. The best lter cascades improve lexicon quality by up to 137% over the plain vanilla statistical method, and approach human performance. Drastically reducing the size of the training corpus has a much smaller impact on lexicon quality when these knowledge sources are used. This makes it practical to train on small hand-built corpora for language pairs where large bilingual corpora are unavailable. Moreover, three of the four lters prove useful even when used with large training corpora. | Automatic Evaluation and Uniform Filter Cascades for Inducing N-Best Translation Lexicons |
d1721388 | Hate speech in the form of racist and sexist remarks are a common occurrence on social media. For that reason, many social media services address the problem of identifying hate speech, but the definition of hate speech varies markedly and is largely a manual effort(BBC, 2015;Lomas, 2015).We provide a list of criteria founded in critical race theory, and use them to annotate a publicly available corpus of more than 16k tweets. We analyze the impact of various extra-linguistic features in conjunction with character n-grams for hatespeech detection. We also present a dictionary based the most indicative words in our data. | Hateful Symbols or Hateful People? Predictive Features for Hate Speech Detection on Twitter |
d14275144 | We consider the task of predicting lexical entailment using distributional vectors. We perform a novel qualitative analysis of one existing model which was previously shown to only measure the prototypicality of word pairs. We find that the model strongly learns to identify hypernyms using Hearst patterns, which are well known to be predictive of lexical relations. We present a novel model which exploits this behavior as a method of feature extraction in an iterative procedure similar to Principal Component Analysis. Our model combines the extracted features with the strengths of other proposed models in the literature, and matches or outperforms prior work on multiple data sets. | Relations such as Hypernymy: Identifying and Exploiting Hearst Patterns in Distributional Vectors for Lexical Entailment |
d12637374 | Even though a linguistics-free sequence to sequence model in neural machine translation (NMT) has certain capability of implicitly learning syntactic information of source sentences, this paper shows that source syntax can be explicitly incorporated into NMT effectively to provide further improvements. Specifically, we linearize parse trees of source sentences to obtain structural label sequences. On the basis, we propose three different sorts of encoders to incorporate source syntax into NMT: 1) Parallel RNN encoder that learns word and label annotation vectors parallelly; 2) Hierarchical RNN encoder that learns word and label annotation vectors in a two-level hierarchy; and 3) Mixed RNN encoder that stitchingly learns word and label annotation vectors over sequences where words and labels are mixed. Experimentation on Chinese-to-English translation demonstrates that all the three proposed syntactic encoders are able to improve translation accuracy. It is interesting to note that the simplest RNN encoder, i.e., Mixed RNN encoder yields the best performance with an significant improvement of 1.4 BLEU points. Moreover, an in-depth analysis from several perspectives is provided to reveal how source syntax benefits NMT. and Charniak, 2016) do. However, the performance gap is very small by adding the ending brackets or not. | Modeling Source Syntax for Neural Machine Translation |
d14081838 | Conventional dependency parsers rely on a statistical model and a transition system or graph algorithm to enforce tree-structured outputs during training and inference.In this work we formalize dependency parsing as the problem of selecting the head (a.k.a.parent) of each word in a sentence.Our model which we call DENSE (as shorthand for Dependency Neural Selection) employs bidirectional recurrent neural networks for the head selection task.Without enforcing any structural constraints during training, DENSE generates (at inference time) trees for the overwhelming majority of sentences (95% on an English dataset), while remaining non-tree outputs can be adjusted with a maximum spanning tree algorithm.We evaluate DENSE on four languages (English, Chinese, Czech, and German) with varying degrees of non-projectivity.Despite the simplicity of our approach, experiments show that the resulting parsers are on par with the state of the art. | Dependency Parsing as Head Selection |
d8634113 | Unknown lexical items present a major obstacle to the development of broadcoverage semantic role labeling systems. We address this problem with a semisupervised learning approach which acquires training instances for unseen verbs from an unlabeled corpus. Our method relies on the hypothesis that unknown lexical items will be structurally and semantically similar to known items for which annotations are available. Accordingly, we represent known and unknown sentences as graphs, formalize the search for the most similar verb as a graph alignment problem and solve the optimization using integer linear programming. Experimental results show that role labeling performance for unknown lexical items improves with training data produced automatically by our method. | Graph Alignment for Semi-Supervised Semantic Role Labeling |
d690568 | We introduce the cross-match test -an exact, distribution free, high-dimensional hypothesis test as an intrinsic evaluation metric for word embeddings. We show that cross-match is an effective means of measuring distributional similarity between different vector representations and of evaluating the statistical significance of different vector embedding models. Additionally, we find that cross-match can be used to provide a quantitative measure of linguistic similarity for selecting bridge languages for machine translation. We demonstrate that the results of the hypothesis test align with our expectations and note that the framework of two sample hypothesis testing is not limited to word embeddings and can be extended to all vector representations. | Hypothesis Testing based Intrinsic Evaluation of Word Embeddings |
d262884128 | Programme Committee Conference Secretariat | |
d263883936 | For machine translation, a vast majority of language pairs in the world are considered low-resource because they have little parallel data available. Besides the technical challenges of learning with limited supervision, it is difficult to evaluate methods trained on lowresource language pairs because of the lack of freely and publicly available benchmarks. In this work, we introduce the FLORES evaluation datasets for Nepali-English and Sinhala-English, based on sentences translated from Wikipedia. Compared to English, these are languages with very different morphology and syntax, for which little out-of-domain parallel data is available and for which relatively large amounts of monolingual data are freely available. We describe our process to collect and cross-check the quality of translations, and we report baseline performance using several learning settings: fully supervised, weakly supervised, semi-supervised, and fully unsupervised. Our experiments demonstrate that current state-of-the-art methods perform rather poorly on this benchmark, posing a challenge to the research community working on lowresource MT. Data and code to reproduce our experiments are available at https://github. com/facebookresearch/flores. | The FLORES Evaluation Datasets for Low-Resource Machine Translation: Nepali-English and Sinhala-English |
d202719215 | Though languages can evolve slowly, they can also react strongly to dramatic world events. By studying the connection between words and events, it is possible to identify which events change our vocabulary and in what way. In this work, we tackle the task of creating timelines-records of historical 'turning points', represented by either words or events, to understand the dynamics of a target word. Our approach identifies these points by leveraging both static and time-varying word embeddings to measure the influence of words and events. In addition to quantifying changes, we show how our technique can help isolate semantic changes. Our qualitative and quantitative evaluations show that we are able to capture this semantic change and event influence. | Generating Timelines by Modeling Semantic Change |
d263896140 | This paper describes an initial prototype demonstrator of a Companion, designed as a platform for novel approaches to the following: 1) The use of Information Extraction (IE) techniques to extract the content of incoming dialogue utterances after an Automatic Speech Recognition (ASR) phase, 2) The conversion of the input to Resource Descriptor Format (RDF) to allow the generation of new facts from existing ones, under the control of a Dialogue Manger (DM), that also has access to stored knowledge and to open knowledge accessed in real time from the web, all in RDF form, 3) A DM implemented as a stack and network virtual machine that models mixed initiative in dialogue control, and 4) A tuned dialogue act detector based on corpus evidence. The prototype platform was evaluated, and we describe this briefly; it is also designed to support more extensive forms of emotion detection carried by both speech and lexical content, as well as extended forms of machine learning. | Demonstration of a prototype for a Conversational Companion for reminiscing about images |
d2616110 | We present a token-level decision summarization framework that utilizes the latent topic structures of utterances to identify "summaryworthy" words.Concretely, a series of unsupervised topic models is explored and experimental results show that fine-grained topic models, which discover topics at the utterance-level rather than the document-level, can better identify the gist of the decisionmaking process. Moreover, our proposed token-level summarization approach, which is able to remove redundancies within utterances, outperforms existing utterance ranking based summarization methods. Finally, context information is also investigated to add additional relevant information to the summary. | Unsupervised Topic Modeling Approaches to Decision Summarization in Spoken Meetings |
d258378314 | Representation of coreferential relations is a challenging and actively studied topic for prodrop and morphologically rich languages (PD-MRLs) due to dropped pronouns (e.g., null subjects and omitted possessive pronouns). These phenomena require a representation scheme at the morphology level and enhanced evaluation methods. In this paper, we propose a representation & evaluation scheme to incorporate dropped pronouns into coreference resolution and validate it on the Turkish language. Using the scheme, we extend the annotations on the only existing Turkish coreference dataset, which originally did not contain annotations for dropped pronouns. We provide publicly available pre and post processors to enhance the prominent CoNLL coreference scorer also to cover coreferential relations arising from dropped pronouns. As a final step, the paper reports the first neural Turkish coreference resolution results in the literature. Although validated on Turkish, the proposed scheme is languageindependent and may be used for other PD-MRLs. 14 | Incorporating Dropped Pronouns into Coreference Resolution: The case for Turkish |
d258378340 | Explanation prompts ask language models to not only assign a label to a given input, such as entailment or contradiction in natural language inference (NLI) tasks, but also to generate a free-text explanation that supports this label. While explanation prompts originally introduced aiming to improve model interpretability, here we show that they also improve robustness to superficial cues. Compared to prompting for labels only, explanation prompting shows stronger performance on adversarial NLI benchmarks, outperforming the state of the art on ANLI, Counterfactually-Augmented NLI, and SNLI-Hard datasets. Analysis suggests that the increase in robustness is due to a reduction in the association strength between single tokens and labels, i.e., explanation prompting weakens superficial cues. More specifically, we find that single tokens that are highly predictive of the correct answer in the label-only setting become uninformative when the model also has to generate explanations. . 2021. Multitask prompted training enables zero-shot task generalization. | Prompting for explanations improves Adversarial NLI. Is this true? {Yes} it is {true} because {it weakens superficial cues} |
d6820419 | In this work, we study parameter tuning towards the M 2 metric, the standard metric for automatic grammar error correction (GEC) tasks. After implementing M 2 as a scorer in the Moses tuning framework, we investigate interactions of dense and sparse features, different optimizers, and tuning strategies for the CoNLL-2014 shared task. We notice erratic behavior when optimizing sparse feature weights with M 2 and offer partial solutions. We find that a bare-bones phrase-based SMT setup with task-specific parameter-tuning outperforms all previously published results for the CoNLL-2014 test set by a large margin (46.37% M 2 over previously 41.75%, by an SMT system with neural features) while being trained on the same, publicly available data. Our newly introduced dense and sparse features widen that gap, and we improve the state-of-the-art to 49.49% M 2 . | Phrase-based Machine Translation is State-of-the-Art for Automatic Grammatical Error Correction |
d263123363 | State-of-the-art statistical approaches to the Coreference Resolution task rely on sophisticated modeling, but very few (10-20) simple features. In this paper we propose to extend the standard feature set substantially, incorporating more linguistic knowledge. To investigate the usability of linguistically motivated features, we evaluate our system for a variety of machine learners on the standard dataset (MUC-7) with the traditional learning set-up (Soon et al., 2001). | Coreference Resolution with and without Linguistic Knowledge |
d130265 | This paper proposes an efficient example selection method for example-based word sense disambiguation systems. To construct a practical size database, a considerable overhead for manual sense disambiguation is required. Our method is characterized by the reliance on the notion of the training utility: the degree to which each example is informative for future example selection when used for the training of the system. The system progressively collects examples by selecting those with greatest utility. The paper reports the effectivity of our method through experiments on about one thousand sentences. Compared to experiments with random example selection, our method reduced the overhead without the degeneration of the performance of the system. | Selective Sampling of Effective Example Sentence Sets for Word Sense Disambiguation |
d263797079 | This paper describes the monomodal and multimodal Neural Machine Translation systems developed by LIUM and CVC for WMT17 Shared Task on Multimodal Translation.We mainly explored two multimodal architectures where either global visual features or convolutional feature maps are integrated in order to benefit from visual context.Our final systems ranked first for both En→De and En→Fr language pairs according to the automatic evaluation metrics METEOR and BLEU. | LIUM-CVC Submissions for WMT17 Multimodal Translation Task |
d170945555 | L'objectif de cet article est d'évaluer dans quelle mesure les "fonctions syntaxiques" qui figurent dans une partie du corpus arboré de Paris 7 sont apprenables à partir d'exemples. La technique d'apprentissage automatique employée pour cela fait appel aux "Champs Aléatoires Conditionnels" (Conditional Random Fields ou CRF), dans une variante adaptée à l'annotation d'arbres. Les expériences menées sont décrites en détail et analysées. Moyennant un bon paramétrage, elles atteignent une F1-mesure de plus de 80%.Abstract. The purpose of this paper is to evaluate whether the "syntactic functions" present in a part of the Paris 7 Treebank are learnable from examples. The learning technic used is the one of "Conditional Random Fields" (CRF), in an original variant adapted to tree labelling. The conducted experiments are extensively described and analyzed. With good parameters, a F1-mesure value of over 80% is reached. | Annotation fonctionnelle de corpus arborés avec des Champs Aléatoires Conditionnels * |
d44086626 | This paper presents a method of designing specific high-order dependency factor on the linear chain conditional random fields (CRFs) for named entity recognition (NER). Named entities tend to be separated from each other by multiple outside tokens in a text, and thus the first-order CRF as well as the second-order CRF may innately lose transition information between distant named entities. The proposed design uses outside label in NER as a transmission medium of precedent entity information on the CRF. Then, empirical results apparently demonstrate that it is possible to exploit long-distance label dependency in the original first-order linear chain CRF structure upon NER while reducing computational loss rather than in the second-order CRF. | Connecting Distant Entities with Induction through Conditional Random Fields for Named Entity Recognition: Precursor-Induced CRF |
d10685747 | This article presents the results we obtained on a complex annotation task (that of dependency syntax) using a specifically designed Game with a Purpose, ZombiLingo. 1 We show that with suitable mechanisms (decomposition of the task, training of the players and regular control of the annotation quality during the game), it is possible to obtain annotations whose quality is significantly higher than that obtainable with a parser, provided that enough players participate. The source code of the game and the resulting annotated corpora (for French) are freely available. This work is licenced under a Creative Commons Attribution 4.0 International License. License details: http:// creativecommons.org/licenses/by/4.0/ 2 See (Church, 2011) for an in-depth reflection on the subject. | Crowdsourcing Complex Language Resources: Playing to Annotate Dependency Syntax |
d49657219 | The goal of our industrial ticketing system is to retrieve a relevant solution for an input query, by matching with historical tickets stored in knowledge base. A query is comprised of subject and description, while a historical ticket consists of subject, description and solution. To retrieve a relevant solution, we use textual similarity paradigm to learn similarity in the query and historical tickets. The task is challenging due to significant term mismatch in the query and ticket pairs of asymmetric lengths, where subject is a short text but description and solution are multi-sentence texts. We present a novel Replicated Siamese LSTM model to learn similarity in asymmetric text pairs, that gives 22% and 7% gain (Accuracy@10) for retrieval task, respectively over unsupervised and supervised baselines. We also show that the topic and distributed semantic features for short and long texts improved both similarity learning and retrieval. This work is licenced under a Creative Commons Attribution 4.0 International Licence. Licence details: http:// creativecommons.org/licenses/by/4.0/ | Replicated Siamese LSTM in Ticketing System for Similarity Learning and Retrieval in Asymmetric Texts |
d177600 | By strictest interpretation, theories of both centering and intonational meaning fail to predict the existence of pitch accented pronominals. Yet they occur felicitously in spoken discourse. To explain this, I emphasize the dual functions served by pitch accents, as markers of both propositional (semantic/pragmatic) and attentional salience. This distinction underlies my proposals about the attentional consequences of pitch accents when applied to pronominals, in particular, that while most pitch accents may weaken or reinforce a cospecifier's status as the center of attention, a contrastively stressed pronominal may force a shift, even when contraindicated by textual features. | The Effect of Pitch Accenting on Pronoun Referent Resolution |
d652921 | Chinese poetry generation is a very challenging task in natural language processing. In this paper, we propose a novel two-stage poetry generating method which first plans the sub-topics of the poem according to the user's writing intent, and then generates each line of the poem sequentially, using a modified recurrent neural network encoder-decoder framework. The proposed planningbased method can ensure that the generated poem is coherent and semantically consistent with the user's intent. A comprehensive evaluation with human judgments demonstrates that our proposed approach outperforms the state-of-the-art poetry generating methods and the poem quality is somehow comparable to human poets. | Chinese Poetry Generation with Planning based Neural Network |
d21701127 | Motivated by a project to create a system for people who are deaf or hard-of-hearing that would use automatic speech recognition (ASR) to produce real-time text captions of spoken English during in-person meetings with hearing individuals, we have augmented a transcript of the Switchboard conversational dialogue corpus with an overlay of word-importance annotations, with a numeric score for each word, to indicate its importance to the meaning of each dialogue turn. Further, we demonstrate the utility of this corpus by training an automatic word importance labeling model; our best performing model has an F-score of 0.60 in an ordinal 6-class word-importance classification task with an agreement (concordance correlation coefficient) of 0.839 with the human annotators (agreement score between annotators is 0.89). Finally, we discuss our intended future applications of this resource, particularly for the task of evaluating ASR performance, i.e. creating metrics that predict ASR-output caption text usability for DHH users better than Word Error Rate (WER). | A Corpus for Modeling Word Importance in Spoken Dialogue Transcripts |
d6376814 | Recent work has attempted to characterize the structure of semantic memory and the search algorithms which, together, best approximate human patterns of search revealed in a semantic fluency task. There are a number of models that seek to capture semantic search processes over networks, but they vary in the cognitive plausibility of their implementation. Existing work has also neglected to consider the constraints that the incremental process of language acquisition must place on the structure of semantic memory. Here we present a model that incrementally updates a semantic network, with limited computational steps, and replicates many patterns found in human semantic fluency using a simple random walk. We also perform thorough analyses showing that a combination of both structural and semantic features are correlated with human performance patterns. | Predicting and Explaining Human Semantic Search in a Cognitive Model |
d11528356 | The quality of statistical machine translation performed with phrase based approaches can be increased by permuting the words in the source sentences in an order which resembles that of the target language. We propose a class of recurrent neural models which exploit source-side dependency syntax features to reorder the words into a target-like order. We evaluate these models on the German-to-English and Italian-to-English language pairs, showing significant improvements over a phrasebased Moses baseline. We also compare with state of the art German-to-English pre-reordering rules, showing that our method obtains similar or better results. | Non-projective Dependency-based Pre-Reordering with Recurrent Neural Network for Machine Translation |
d54434776 | We propose a novel method, Modality-based Redundancy Reduction Fusion (MRRF), for understanding and modulating the relative contribution of each modality in multimodal inference tasks. This is achieved by obtaining an (M + 1)-way tensor to consider the high-order relationships between M modalities and the output layer of a neural network model. Applying a modality-based tensor factorization method, which adopts different factors for different modalities, results in removing information present in a modality that can be compensated by other modalities, with respect to model outputs. This helps to understand the relative utility of information in each modality. In addition it leads to a less complicated model with less parameters and therefore could be applied as a regularizer avoiding overfitting. We have applied this method to three different multimodal datasets in sentiment analysis, personality trait recognition, and emotion recognition. We are able to recognize relationships and relative importance of different modalities in these tasks and achieves a 1% to 4% improvement on several evaluation measures compared to the state-of-the-art for all three tasks. | Modality-based Factorization for Multimodal Fusion |
d34170652 | 率 句 索 立 | |
d258486848 | This paper describes the structure and findings of the SIGTYP 2023 shared task on cognate and derivative detection for low-resourced languages, broken down into a supervised and unsupervised sub-task. The participants were asked to submit the test data's final prediction. A total of nine teams registered for the shared task where seven teams registered for both subtasks. Only two participants ended up submitting system descriptions, with only one submitting systems for both sub-tasks. While all systems show a rather promising performance, all could be within the baseline score for the supervised sub-task. However, the system submitted for the unsupervised sub-task outperforms the baseline score. | Findings of the SIGTYP 2023 Shared task on Cognate and Derivative Detection For Low-Resourced Languages |
d9519654 | Morphological segmentation breaks words into morphemes (the basic semantic units). It is a key component for natural language processing systems. Unsupervised morphological segmentation is attractive, because in every language there are virtually unlimited supplies of text, but very few labeled resources. However, most existing model-based systems for unsupervised morphological segmentation use directed generative models, making it difficult to leverage arbitrary overlapping features that are potentially helpful to learning. In this paper, we present the first log-linear model for unsupervised morphological segmentation. Our model uses overlapping features such as morphemes and their contexts, and incorporates exponential priors inspired by the minimum description length (MDL) principle. We present efficient algorithms for learning and inference by combining contrastive estimation with sampling. Our system, based on monolingual features only, outperforms a state-of-the-art system by a large margin, even when the latter uses bilingual information such as phrasal alignment and phonetic correspondence. On the Arabic Penn Treebank, our system reduces F1 error by 11% compared to Morfessor. | Unsupervised Morphological Segmentation with Log-Linear Models |
d9928714 | In this work, we improve the performance of intra-sentential zero anaphora resolution in Japanese using a novel method of recognizing subject sharing relations. In Japanese, a large portion of intrasentential zero anaphora can be regarded as subject sharing relations between predicates, that is, the subject of some predicate is also the unrealized subject of other predicates. We develop an accurate recognizer of subject sharing relations for pairs of predicates in a single sentence, and then construct a subject shared predicate network, which is a set of predicates that are linked by the subject sharing relations recognized by our recognizer. We finally combine our zero anaphora resolution method exploiting the subject shared predicate network and a state-ofthe-art ILP-based zero anaphora resolution method. Our combined method achieved a significant improvement over the the ILPbased method alone on intra-sentential zero anaphora resolution in Japanese. To the best of our knowledge, this is the first work to explicitly use an independent subject sharing recognizer in zero anaphora resolution. | Intra-sentential Zero Anaphora Resolution using Subject Sharing Recognition |
d44091475 | Natural language generation (NLG) is a critical component in spoken dialogue systems. Classic NLG can be divided into two phases:(1) sentence planning: deciding on the overall sentence structure, (2) surface realization: determining specific word forms and flattening the sentence structure into a string. Many simple NLG models are based on recurrent neural networks (RNN) and sequence-to-sequence (seq2seq) model, which basically contains a encoder-decoder structure; these NLG models generate sentences from scratch by jointly optimizing sentence planning and surface realization using a simple cross entropy loss training criterion. However, the simple encoderdecoder architecture usually suffers from generating complex and long sentences, because the decoder has to learn all grammar and diction knowledge. This paper introduces a hierarchical decoding NLG model based on linguistic patterns in different levels, and shows that the proposed method outperforms the traditional one with a smaller model size. Furthermore, the design of the hierarchical decoding is flexible and easily-extensible in various NLG systems 1 . | Natural Language Generation by Hierarchical Decoding with Linguistic Patterns |
d28217923 | This paper revisits the problem of complex word identification (CWI) following up the SemEval CWI shared task. We use ensemble classifiers to investigate how well computational methods can discriminate between complex and noncomplex words. Furthermore, we analyze the classification performance to understand what makes lexical complexity challenging. Our findings show that most systems performed poorly on the SemEval CWI dataset, and one of the reasons for that is the way in which human annotation was performed. | Complex Word Identification: Challenges in Data Annotation and System Performance |
d6166308 | We adapt the "hook" trick for speeding up bilexical parsing to the decoding problem for machine translation models that are based on combining a synchronous context free grammar as the translation model with an n-gram language model. This dynamic programming technique yields lower complexity algorithms than have previously been described for an important class of translation models.where e is a source language word, f is a foreign language word, and means the null token, and binary production rules in two forms that are responsible for generating syntactic subtree pairs: | Machine Translation as Lexicalized Parsing with Hooks |
d263802845 | This paper provides an algebraic characterization of the total input strictly local functions.Simultaneous, noniterative rules of the form A→B/C D, common in phonology, are definable as functions in this class whenever CAD represents a finite set of strings.The algebraic characterization highlights a fundamental connection between input strictly local functions and the simple class of definite string languages, as well as connections to string functions studied in the computer science literature, the definite functions and local functions.No effective decision procedure for the input strictly local maps was previously available, but one arises directly from this characterization.This work also shows that, unlike the full class, a restricted subclass is closed under composition.Additionally, some products are defined which may yield new factorization methods. | An Algebraic Characterization of Total Input Strictly Local Functions |
d3986974 | Answering complex questions is a timeconsuming activity for humans that requires reasoning and integration of information. Recent work on reading comprehension made headway in answering simple questions, but tackling complex questions is still an ongoing research challenge. Conversely, semantic parsers have been successful at handling compositionality, but only when the information resides in a target knowledge-base. In this paper, we present a novel framework for answering broad and complex questions, assuming answering simple questions is possible using a search engine and a reading comprehension model. We propose to decompose complex questions into a sequence of simple questions, and compute the final answer from the sequence of answers. To illustrate the viability of our approach, we create a new dataset of complex questions, COMPLEXWEBQUES-TIONS, and present a model that decomposes questions and interacts with the web to compute an answer. We empirically demonstrate that question decomposition improves performance from 20.8 precision@1 to 27.5 preci-sion@1 on this new dataset. | The Web as a Knowledge-base for Answering Complex Questions |
d11994461 | Crowdsourcing, which offers new ways of cheaply and quickly gathering large amounts of information contributed by volunteers online, has revolutionised the collection of labelled data. Yet, to create annotated linguistic resources from this data, we face the challenge of having to combine the judgements of a potentially large group of annotators. In this paper we investigate how to aggregate individual annotations into a single collective annotation, taking inspiration from the field of social choice theory. We formulate a general formal model for collective annotation and propose several aggregation methods that go beyond the commonly used majority rule. We test some of our methods on data from a crowdsourcing experiment on textual entailment annotation. | Collective Annotation of Linguistic Resources: Basic Principles and a Formal Model |
d151531619 | La prise en compte des informations auditives et proprioceptives dans le contrôle de la parole est mise en évidence par un nombre croissant de résultats expérimentaux. Cependant, les modèles de production imposent le plus souvent l'une ou l'autre des modalités, ou n'offrent pas de cadre formel pour évaluer leurs contributions respectives. Nous proposons d'explorer le rôle de ces modalités sensorielles dans la planification des gestes de parole à partir d'un modèle bayésien représentant la structure des connaissances mises en jeu dans cette tâche. Le modèle permet d'envisager trois mécanismes de planification, reposant sur la modalité auditive, proprioceptive ou sur les deux conjointement. Nous comparons des simulations obtenues par les deux premiers mécanismes de planification. Les résultats indiquent des réalisations articulatoires différentes mais donnant néanmoins des réalisations auditives qualitativement similaires dans leur variabilité.ABSTRACTBayesian modeling of speech gesture motor planning: Evaluating the role of different sensory modalities An increasing number of experimental results have identified a clear role of auditory and somatosensory information in speech motor control. However, most of the speech production models consider only one of these sensory modalities, or do not provide the possibility to formally evaluate the respective contribution of these modalities. We propose to explore the role of auditory and proprioceptive representations in speech gesture planning, based on a Bayesian model representing the structure of knowledge involved. The model allows to consider three planning mechanisms, based on the auditory or proprioceptive modality or the combination of both. We compare simulations obtained from the two first planning mechanisms. Results indicate differences in the generated articulatory patterns, giving rise however to qualitatively similar patterns of auditory variability. MOTS-CLÉS : Contrôle moteur de la parole -Modélisation bayésienne -Multimodalité . | Modélisation bayésienne de la planification motrice des gestes de parole : Évaluation du rôle des différentes modalités sensorielles |
d2711033 | We present the results of feature engineering and post-processing experiments conducted on a temporal expression recognition task. The former explores the use of different kinds of tagging schemes and of exploiting a list of core temporal expressions during training. The latter is concerned with the use of this list for postprocessing the output of a system based on conditional random fields.We find that the incorporation of knowledge sources both for training and postprocessing improves recall, while the use of extended tagging schemes may help to offset the (mildly) negative impact on precision. Each of these approaches addresses a different aspect of the overall recognition performance. Taken separately, the impact on the overall performance is low, but by combining the approaches we achieve both high precision and high recall scores. | Feature Engineering and Post-Processing for Temporal Expression Recognition Using Conditional Random Fields |
d198986161 | The CMU Wilderness Multilingual Speech Dataset (Black, 2019) is a newly published multilingual speech dataset based on recorded readings of the New Testament. It provides data to build Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) models for potentially 700 languages. However, the fact that the source content (the Bible) is the same for all the languages is not exploited to date. Therefore, this article proposes to add multilingual links between speech segments in different languages, and shares a large and clean dataset of 8,130 parallel spoken utterances across 8 languages (56 language pairs). We name this corpus MaSS (Multilingual corpus of Sentence-aligned Spoken utterances). The covered languages (Basque, English, Finnish, French, Hungarian, Romanian, Russian and Spanish) allow researches on speech-to-speech alignment as well as on translation for typologically different language pairs. The quality of the final corpus is attested by human evaluation performed on a corpus subset (100 utterances, 8 language pairs). Lastly, we showcase the usefulness of the final product on a bilingual speech retrieval task. | MaSS: A Large and Clean Multilingual Corpus of Sentence-aligned Spoken Utterances Extracted from the Bible |
d53047545 | We introduce a method to reduce constituent parsing to sequence labeling. For each word w t , it generates a label that encodes: (1) the number of ancestors in the tree that the words w t and w t+1 have in common, and (2) the nonterminal symbol at the lowest common ancestor. We first prove that the proposed encoding function is injective for any tree without unary branches. In practice, the approach is made extensible to all constituency trees by collapsing unary branches. We then use the PTB and CTB treebanks as testbeds and propose a set of fast baselines. We achieve 90% F-score on the PTB test set, outperforming the Vinyals et al. (2015) sequence-to-sequence parser. In addition, sacrificing some accuracy, our approach achieves the fastest constituent parsing speeds reported to date on PTB by a wide margin. | Constituent Parsing as Sequence Labeling |
d21726416 | The TERRE-ISTEX project aims to identify scientific research dealing with specific geographical territories areas based on heterogeneous digital content available in scientific papers. The project is divided into three main work packages: (1) identification of the periods and places of empirical studies, and which reflect the publications resulting from the analyzed text samples, (2) identification of the themes which appear in these documents, and (3) development of a web-based geographical information retrieval tool (GIR). The first two actions combine Natural Language Processing patterns with text mining methods. The integration of the spatial, thematic and temporal dimensions in a GIR contributes to a better understanding of what kind of research has been carried out, of its topics and its geographical and historical coverage. Another originality of the TERRE-ISTEX project is the heterogeneous character of the corpus, including PhD theses and scientific articles from the ISTEX digital libraries and the CIRAD research center. | Automatic Identification of Research Fields in Scientific Papers |
d8764466 | We demonstrate significant improvement on the MCTest question answering task(Richardson et al., 2013)by augmenting baseline features with features based on syntax, frame semantics, coreference, and word embeddings, and combining them in a max-margin learning framework. We achieve the best results we are aware of on this dataset, outperforming concurrentlypublished results. These results demonstrate a significant performance gradient for the use of linguistic structure in machine comprehension. | Machine Comprehension with Syntax, Frames, and Semantics |
d44924568 | Nous présentons dans cet article un travail portant sur la création d'un corpus de français parlé spontané annoté en morphosyntaxe. Nous détaillons la méthodologie suivie afin d'assurer le contrôle de la qualité de la ressource finale. Ce corpus est d'ores et déjà librement diffusé pour la recherche et peut servir aussi bien de corpus d'apprentissage pour des logiciels que de base pour des descriptions linguistiques. Nous présentons également les résultats obtenus par deux étiqueteurs morphosyntaxiques entrainés sur ce corpus.ABSTRACTTCOF-POS : A Freely Available POS-Tagged Corpus of Spoken FrenchThis article details the creation of TCOF-POS, the first freely available corpus of spontaneous spoken French. We present here the methodology that was followed in order to obtain the best possible quality in the final resource. This corpus already is freely available and can be used as a training/validation corpus for NLP tools, as well as a study corpus for linguistic research. We also present the results obtained by two POS-taggers trained on the corpus. MOTS-CLÉS : Etiquetage morpho-syntaxique, français parlé, ressources langagières. | TCOF-POS : un corpus libre de français parlé annoté en morphosyntaxe |
d67788603 | Despite recent advances in natural language processing, many statistical models for processing text perform extremely poorly under domain shift. Processing biomedical and clinical text is a critically important application area of natural language processing, for which there are few robust, practical, publicly available models. This paper describes scis-paCy, a new Python library and models for practical biomedical/scientific text processing, which heavily leverages the spaCy library. We detail the performance of two packages of models released in scispaCy and demonstrate their robustness on several tasks and datasets. Models and code are available at https:// allenai.github.io/scispacy/. | ScispaCy: Fast and Robust Models for Biomedical Natural Language Processing |
d46894722 | Building tools for code-mixed data is rapidly gaining popularity in the NLP research community as such data is exponentially rising on social media. Working with code-mixed data contains several challenges, especially due to grammatical inconsistencies and spelling variations in addition to all the previous known challenges for social media scenarios. In this article, we present a novel architecture focusing on normalizing phonetic typing variations, which is commonly seen in code-mixed data. One of the main features of our architecture is that in addition to normalizing, it can also be utilized for back-transliteration and word identification in some cases. Our model achieved an accuracy of 90.27% on the test data. | Normalization of Transliterated Words in Code-Mixed Data Using Seq2Seq Model & Levenshtein Distance |
d207847794 | ||
d1397 | Ira this paper I present a parser based on Description Logics (I)I.) for a German lll'SG-slyle fragment. The specilied parser relies mainly on the inferential capabilities of the nnderlying DL system. Given a preferential default extension for DL disamhiguation is achieved by choosing the parse containing a qualitatively minimal number of exceptions. | An HPSG Parser Based on Description l,ogies* |
d202537185 | Language technologies play a key role in assisting people with their writing. Although there has been steady progress in e.g., grammatical error correction (GEC), human writers are yet to benefit from this progress due to the high development cost of integrating with writing software. We propose TEASPN 1 , a protocol and an open-source framework for achieving integrated writing assistance environments. The protocol standardizes the way writing software communicates with servers that implement such technologies, allowing developers and researchers to integrate the latest developments in natural language processing (NLP) with low cost. As a result, users can enjoy the integrated experience in their favorite writing software. The results from experiments with human participants show that users use a wide range of technologies and rate their writing experience favorably, allowing them to write more fluent text.BackendResearch ModelsWriting Software Writing Software | TEASPN: Framework and Protocol for Integrated Writing Assistance Environments |
d263897670 | In order to alleviate data sparsity and overfitting problems in maximum likelihood estimation (MLE) for sequence prediction tasks, we propose the Generative Bridging Network (GBN), in which a novel bridge module is introduced to assist the training of the sequence prediction model (the generator network). Unlike MLE directly maximizing the conditional likelihood, the bridge extends the point-wise ground truth to a bridge distribution conditioned on it, and the generator is optimized to minimize their KL-divergence. Three different GBNs, namely uniform GBN, language-model GBN and coaching GBN, are proposed to penalize confidence, enhance language smoothness and relieve learning burden. Experiments conducted on two recognized sequence prediction tasks (machine translation and abstractive text summarization) show that our proposed GBNs can yield significant improvements over strong baselines. Furthermore, by analyzing samples drawn from different bridges, expected influences on the generator are verified. | Generative Bridging Network for Neural Sequence Prediction |
d1509090 | Multiword expressions (MWEs) are a class of linguistic forms spanning conventional word boundaries that are both idiosyncratic and pervasive across different languages. The structure of linguistic processing that depends on the clear distinction between words and phrases has to be re-thought to accommodate MWEs. The issue of MWE handling is crucial for NLP applications, where it raises a number of challenges. The emergence of solutions in the absence of guiding principles motivates this survey, whose aim is not only to provide a focused review Volume 43, Number 4 of MWE processing, but also to clarify the nature of interactions between MWE processing and downstream applications. We propose a conceptual framework within which challenges and research contributions can be positioned. It offers a shared understanding of what is meant by "MWE processing," distinguishing the subtasks of MWE discovery and identification. It also elucidates the interactions between MWE processing and two use cases: Parsing and machine translation. Many of the approaches in the literature can be differentiated according to how MWE processing is timed with respect to underlying use cases. We discuss how such orchestration choices affect the scope of MWE-aware systems. For each of the two MWE processing subtasks and for each of the two use cases, we conclude on open issues and research perspectives. r Clearly define the task of MWE processing and delineate its two main subtasks, discovery and identification, r Elaborate on the interaction between MWE processing and two selected NLP use cases, that is, parsing and MT, r Explain how the key MWE properties, such as discontiguity, variability, and non-compositionality, give rise to challenges and opportunities for MWE processing and MWE-aware applications such as parsing and MT.r Differentiate previous work according to how MWE processing is timed ("orchestrated") with respect to the underlying application. | Multiword Expression Processing: A Survey under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license Computational Linguistics |
d49559642 | Machine translation from polysynthetic to fusional languages is a challenging task, which gets further complicated by the limited amount of parallel text available. Thus, translation performance is far from the state of the art for high-resource and more intensively studied language pairs. To shed light on the phenomena which hamper automatic translation to and from polysynthetic languages, we study translations from three low-resource, polysynthetic languages (Nahuatl, Wixarika and Yorem Nokki) into Spanish and vice versa. Doing so, we find that in a morpheme-to-morpheme alignment an important amount of information contained in polysynthetic morphemes has no Spanish counterpart, and its translation is often omitted. We further conduct a qualitative analysis and, thus, identify morpheme types that are commonly hard to align or ignored in the translation process.This work is licensed under a Creative Commons Attribution 4.0 International License. License details: http:// creativecommons.org/licenses/by/4.0/. | Lost in Translation: Analysis of Information Loss During Machine Translation Between Polysynthetic and Fusional Languages |
d53037336 | We introduce PreCo, a large-scale English dataset for coreference resolution. The dataset is designed to embody the core challenges in coreference, such as entity representation, by alleviating the challenge of low overlap between training and test sets and enabling separated analysis of mention detection and mention clustering. To strengthen the training-test overlap, we collect a large corpus of 38K documents and 12.5M words which are mostly from the vocabulary of Englishspeaking preschoolers. Experiments show that with higher training-test overlap, error analysis on PreCo is more efficient than the one on OntoNotes, a popular existing dataset. Furthermore, we annotate singleton mentions making it possible for the first time to quantify the influence that a mention detector makes on coreference resolution performance. The dataset is freely available at https:// preschool-lab.github.io/PreCo/. | PreCo: A Large-scale Dataset in Preschool Vocabulary for Coreference Resolution |
d16676196 | This paper attempts to provide a new analysis of multiple wh-fragments, the socalled 'multiple sluicing', in English. Against previous approaches which resort to a reconstruction/deletion account or a gapping analysis, the proposed analysis captures multiple wh-fragments without positing a copy-deletion mechanism. In fact, multiple whfragments are viewed as a root clause whose interpretation is determined by previous discourse context. This is achieved by suggesting a new type constraint based on the analysis of fragment byBertomeu and Kordoni (2005). | Resolving Multiple Wh-Fragments: A Syntax-Pragmatics Approach * * * * |
d243865233 | Previous works on syntactically controlled paraphrase generation heavily rely on largescale parallel paraphrase data that are not easily available for many languages and domains. In this paper, we take this research direction to the extreme and investigate whether it is possible to learn syntactically controlled paraphrase generation with non-parallel data. We propose a syntactically-informed unsupervised paraphrasing model based on conditional variational auto-encoder (VAE) which can generate texts in a specified syntactic structure. Particularly, we design a two-stage learning method to effectively train the model using non-parallel data. The conditional VAE is trained to reconstruct the input sentence according to the given input and its syntactic structure. Furthermore, to improve the syntactic controllability and semantic consistency of the pre-trained conditional VAE, we finetune it using syntax controlling and cycle reconstruction learning objectives, and employ Gumbel-Softmax to combine these new learning objectives. Experiment results demonstrate that the proposed model trained only on non-parallel data is capable of generating diverse paraphrases with specified structures. Additionally, we further validate the effectiveness of our method for generating syntactically adversarial examples on a sentiment analysis task. Source codes are available at https: //github.com/lanse-sir/sup. * Corresponding Authors. | Syntactically-Informed Unsupervised Paraphrasing with Non-Parallel Data |
d10720120 | Supervised hierarchical topic modeling and unsupervised hierarchical topic modeling are usually used to obtain hierarchical topics, such as hLLDA and hLDA. Supervised hierarchical topic modeling makes heavy use of the information from observed hierarchical labels, but cannot explore new topics; while unsupervised hierarchical topic modeling is able to detect automatically new topics in the data space, but does not make use of any information from hierarchical labels. In this paper, we propose a semi-supervised hierarchical topic model which aims to explore new topics automatically in the data space while incorporating the information from observed hierarchical labels into the modeling process, called Semi-Supervised Hierarchical Latent Dirichlet Allocation (SSHLDA). We also prove that hLDA and hLLDA are special cases of SSHLDA. We conduct experiments on Yahoo! Answers and ODP datasets, and assess the performance in terms of perplexity and clustering. The experimental results show that predictive ability of SSHLDA is better than that of baselines, and SSHLDA can also achieve significant improvement over baselines for clustering on the FScore measure. | SSHLDA: A Semi-Supervised Hierarchical Topic Model |
d5139774 | Traditional vector-based models use word co-occurrence counts from large corpora to represent lexical meaning. In this paper we present a novel approach for constructing semantic spaces that takes syntactic relations into account. We introduce a formalisation for this class of models and evaluate their adequacy on two modelling tasks: semantic priming and automatic discrimination of lexical relations. | Constructing Semantic Space Models from Parsed Corpora |
d6630816 | Social tagging provides an efficient way to manage online resources. In order to collect more social tags, many research efforts aim to automatically suggest tags to help users annotate tags. Many content-based methods assume tags are independent and suggest tags one by one independently. Although it makes suggestion easier, the independence assumption does not confirm to reality, and the suggested tags are usually inconsistent and incoherent with each other. To address this problem, we propose to model contextaware relations of tags for suggestion: (1) By regarding resource content as context of tags, we propose Tag Context Model to identify specific context words in resource content for tags. (2) Given a new resource, we build a context-aware relation graph of candidate tags, and propose a random walk algorithm to rank tags for suggestion. Experiment results demonstrate our method outperforms other state-of-the-art methods.TITLE AND ABSTRACT IN CHINESE | Random Walks on Context-Aware Relation Graphs for Ranking Social Tags TITLE AND ABSTRACT IN CHINESE |
d202676699 | Systems that can associate images with their spoken audio captions are an important step towards visually grounded language learning. We describe a scalable method to automatically generate diverse audio for image captioning datasets. This supports pretraining deep networks for encoding both audio and images, which we do via a dual encoder that learns to align latent representations from both modalities. We show that a masked margin softmax loss for such models is superior to the standard triplet loss. We fine-tune these models on the Flickr8k Audio Captions Corpus and obtain state-of-the-art results-improving recall in the top 10 from 29.6% to 49.5%. We also obtain human ratings on retrieval outputs to better assess the impact of incidentally matching image-caption pairs that were not associated in the data, finding that automatic evaluation substantially underestimates the quality of the retrieved results. * Work done as a member of the Google AI Residency Program. | Large-scale representation learning from visually grounded untranscribed speech |
d202677204 | Universal Conceptual Cognitive Annotation (UCCA; Abend and Rappoport, 2013) is a typologically-informed, broad-coverage semantic annotation scheme that describes coarse-grained predicate-argument structure but currently lacks semantic roles. We argue that lexicon-free annotation of the semantic roles marked by prepositions, as formulated by Schneider et al.(2018), is complementary and suitable for integration within UCCA. We show empirically for English that the schemes, though annotated independently, are compatible and can be combined in a single semantic graph. A comparison of several approaches to parsing the integrated representation lays the groundwork for future research on this task. | Made for Each Other: Broad-coverage Semantic Structures Meet Preposition Supersenses |
d16086013 | Two extensions to the AMR smatch scoring script are presented. The first extension combines the smatch scoring script with the C6.0 rule-based classifier to produce a human-readable report on the error patterns frequency observed in the scored AMR graphs. This first extension results in 4% gain over the state-of-art CAMR baseline parser by adding to it a manually crafted wrapper fixing the identified CAMR parser errors. The second extension combines a per-sentence smatch with an ensemble method for selecting the best AMR graph among the set of AMR graphs for the same sentence. This second modification automatically yields further 0.4% gain when applied to outputs of two nondeterministic AMR parsers: a CAMR+wrapper parser and a novel character-level neural translation AMR parser. For AMR parsing task the character-level neural translation attains surprising 7% gain over the carefully optimized word-level neural translation. Overall, we achieve smatch F1=62% on the SemEval-2016 official scoring set and F1=67% on the LDC2015E86 test set. | RIGA at SemEval-2016 Task 8: Impact of Smatch Extensions and Character-Level Neural Translation on AMR Parsing Accuracy |
d7637949 | This paper describes the construction of language choice models for the microplanning of discourse relations in a Natural Language Generation system that attempts to generate appropriate texts for users with varying levels of literacy. The models consist of constraint satisfaction problem graphs that have been derived from the results of a corpus analysis. The corpus that the models are based on was written for good readers. We adapted the models for poor readers by allowing certain constraints to be tightened, based on psycholinguistic evidence. We describe how the design of microplanner is evolving. We discuss the compromises involved in generating more readable textual output and implications of our design for NLG architectures. Finally we describe plans for future work. | Language choice models for microplanning and readability |
d5112203 | We present a model for semantic proto-role labeling (SPRL) using an adapted bidirectional LSTM encoding strategy that we call Neural-Davidsonian: predicate-argument structure is represented as pairs of hidden states corresponding to predicate and argument head tokens of the input sequence. We demonstrate:(1) state-of-the-art results in SPRL, and (2) that our network naturally shares parameters between attributes, allowing for learning new attribute types with limited added supervision. The cat ate the rat Word Embeddings BiLSTM Wshared ReLU Wchanged_state Wvolition Wexisted_after hate hrat Neural Davidsonian Semantic Proto-roles changed_state(eate, rat) existed_after(eate, rat) volition(eate, rat) | Neural-Davidsonian Semantic Proto-role Labeling |
d170997589 | Nous définissons le β-calcul, un calcul de réécriture de graphes, que nous proposons d'utiliser pour étudier les liens entre différentes représentations linguistiques. Nous montrons comment transformer une analyse syntaxique en une représentation sémantique par la composition de deux jeux de règles de β-calcul. Le premier souligne l'importance de certaines informations syntaxiques pour le calcul de la sémantique et explicite le lien entre syntaxe et sémantique sous-spécifiée. Le second décompose la recherche de modèles pour les représentations sémantiques sous-spécifiées.Abstract. We define the β-calculus, a graph-rewriting calculus, which we propose to use to study the links between different linguistic representations. We show how to transform a syntactic analysis into a semantic analysis via the composition of two sets of β-calculus rules. The first one underlines the importance of some syntactic information to compute the semantics and clearly expresses the link between syntax and underspecified semantics. The second one breaks up the search for models of underspecified semantic representations.Mots-clés : Dépendances, réécriture de graphes, interface syntaxe-sémantique, DMRS. | Réécriture de graphes de dépendances pour l'interface syntaxe-sémantique |
d1356419 | We present results of two methods for assessing the event profile of news articles as a function of verb type. The unique contribution of this research is the focus on the role of verbs, rather than nouns. Two algorithms are presented and evaluated, one of which is shown to accurately discriminate documents by type and semantic properties, i.e. the event profile. The initial method, using WordNet(Miller et al. 1990), produced multiple cross-classification of articles, primarily due to the bushy nature of the verb tree coupled with the sense disambiguation problem. Our second approach using English Verb Classes and Alternations (EVCA)Levin (1993)showed that monosemous categorization of the frequent verbs in WSJ made it possible to usefully discriminate documents. For example, our results show that articles in which communication verbs predominate tend to be opinion pieces, whereas articles with a high percentage of agreement verbs tend to be about mergers or legal cases. An evaluation is performed on the results using Kendall's ~-. We present convincing evidence for using verb semantic classes as a discriminant in document classification. 1 | Role of Verbs in Document Analysis |
d13746581 | Peer reviewing is a central component in the scientific publishing process. We present the first public dataset of scientific peer reviews available for research purposes (PeerRead v1), 1 providing an opportunity to study this important artifact. The dataset consists of 14.7K paper drafts and the corresponding accept/reject decisions in top-tier venues including ACL, NIPS and ICLR. The dataset also includes 10.7K textual peer reviews written by experts for a subset of the papers. We describe the data collection process and report interesting observed phenomena in the peer reviews. We also propose two novel NLP tasks based on this dataset and provide simple baseline models. In the first task, we show that simple models can predict whether a paper is accepted with up to 21% error reduction compared to the majority baseline. In the second task, we predict the numerical scores of review aspects and show that simple models can outperform the mean baseline for aspects with high variance such as 'originality' and 'impact'. 2 The 20 th SIGNLL Conference on Computational Natural Language Learning; http://www.conll.org/2016 3 The 55 th Annual Meeting of the Association for Computational Linguistics; http://acl2017.org/ 4 The Conference on Neural Information Processing Systems; https://nips.cc/ 5 http://openreview.net 6 The 5 th International Conference on Learning Representations; https://iclr.cc/archive/www/2017.html 7 The platform also allows any person to review the paper by adding a comment, but we only use the official reviews of reviewers assigned to review that paper. 8 https://arxiv.org/ 9 For consistency, we only include the first arXiv version | A Dataset of Peer Reviews (PeerRead): Collection, Insights and NLP Applications |
d10010426 | We investigate the characteristics of factual and emotional argumentation styles observed in online debates. Using an annotated set of FACTUAL and FEELING debate forum posts, we extract patterns that are highly correlated with factual and emotional arguments, and then apply a bootstrapping methodology to find new patterns in a larger pool of unannotated forum posts. This process automatically produces a large set of patterns representing linguistic expressions that are highly correlated with factual and emotional language. Finally, we analyze the most discriminating patterns to better understand the defining characteristics of factual and emotional arguments. | And That's A Fact: Distinguishing Factual and Emotional Argumentation in Online Dialogue |
d53082542 | We propose and compare methods for gradientbased domain adaptation of self-attentive neural machine translation models. We demonstrate that a large proportion of model parameters can be frozen during adaptation with minimal or no reduction in translation quality by encouraging structured sparsity in the set of offset tensors during learning via group lasso regularization. We evaluate this technique for both batch and incremental adaptation across multiple data sets and language pairs. Our system architecture-combining a state-of-the-art self-attentive model with compact domain adaptation-provides high quality personalized machine translation that is both space and time efficient. | Compact Personalized Models for Neural Machine Translation |
d53101509 | Rapidly expanding volume of publications in the biomedical domain makes it increasingly difficult for a timely evaluation of the latest literature. That, along with a push for automated evaluation of clinical reports, present opportunities for effective natural language processing methods. In this study we target the problem of named entity recognition, where texts are processed to annotate terms that are relevant for biomedical studies. Terms of interest in the domain include gene and protein names, and cell lines and types. Here we report on a pipeline built on Embeddings from Language Models (ELMo) and a deep learning package for natural language processing (Al-lenNLP). We trained context-aware token embeddings on a dataset of biomedical papers using ELMo, and incorporated these embeddings in the LSTM-CRF model used by AllenNLP for named entity recognition. We show these representations improve named entity recognition for different types of biomedical named entities. We also achieve a new state of the art in gene mention detection on the BioCreative II gene mention shared task. | In-domain Context-aware Token Embeddings Improve Biomedical Named Entity Recognition |
d14401063 | Neural networks are among the state-ofthe-art techniques for language modeling. Existing neural language models typically map discrete words to distributed, dense vector representations. After information processing of the preceding context words by hidden layers, an output layer estimates the probability of the next word. Such approaches are time-and memory-intensive because of the large numbers of parameters for word embeddings and the output layer. In this paper, we propose to compress neural language models by sparse word representations. In the experiments, the number of parameters in our model increases very slowly with the growth of the vocabulary size, which is almost imperceptible. Moreover, our approach not only reduces the parameter space to a large extent, but also improves the performance in terms of the perplexity measure. 1 | Compressing Neural Language Models by Sparse Word Representations |
d3036949 | Neural networks have been shown to improve performance across a range of natural-language tasks. However, designing and training them can be complicated. Frequently, researchers resort to repeated experimentation to pick optimal settings. In this paper, we address the issue of choosing the correct number of units in hidden layers. We introduce a method for automatically adjusting network size by pruning out hidden units through ∞,1 and 2,1 regularization. We apply this method to language modeling and demonstrate its ability to correctly choose the number of hidden units while maintaining perplexity. We also include these models in a machine translation decoder and show that these smaller neural models maintain the significant improvements of their unpruned versions. | Auto-Sizing Neural Networks: With Applications to n-gram Language Models |
d5101528 | Despite substantial progress made in developing new sentiment lexicon generation (SLG) methods for English, the task of transferring these approaches to other languages and domains in a sound way still remains open. In this paper, we contribute to the solution of this problem by systematically comparing semi-automatic translations of common English polarity lists with the results of the original automatic SLG algorithms, which were applied directly to German data. We evaluate these lexicons on a corpus of 7,992 manually annotated tweets. In addition to that, we also collate the results of dictionary-and corpus-based SLG methods in order to find out which of these paradigms is better suited for the inherently noisy domain of social media. Our experiments show that semi-automatic translations notably outperform automatic systems (reaching a macro-averaged F 1 -score of 0.589), and that dictionary-based techniques produce much better polarity lists as compared to corpus-based approaches (whose best F 1 -scores run up to 0.479 and 0.419 respectively) even for the non-standard Twitter genre. All reimplementations of the compared systems and the resulting lexicons of these methods are available online at https://github.com/WladimirSidorenko/SentiLex. This work is licenced under a Creative Commons Attribution 4.0 International License. License details: | Generating Sentiment Lexicons for German Twitter |
d10549256 | We describe the University of Sheffield system used in the TempEval-2 challenge, USFD2. The challenge requires the automatic identification of temporal entities and relations in text. USFD2 identifies and anchors temporal expressions, and also attempts two of the four temporal relation assignment tasks. A rule-based system picks out and anchors temporal expressions, and a maximum entropy classifier assigns temporal link labels, based on features that include descriptions of associated temporal signal words. USFD2 identified temporal expressions successfully, and correctly classified their type in 90% of cases. Determining the relation between an event and time expression in the same sentence was performed at 63% accuracy, the second highest score in this part of the challenge. | USFD2: Annotating Temporal Expresions and TLINKs for TempEval-2 |
d23144639 | We consider the problem of learning general-purpose, paraphrastic sentence embeddings in the setting ofWieting et al. (2016b). We use neural machine translation to generate sentential paraphrases via back-translation of bilingual sentence pairs. We evaluate the paraphrase pairs by their ability to serve as training data for learning paraphrastic sentence embeddings. We find that the data quality is stronger than prior work based on bitext and on par with manually-written English paraphrase pairs, with the advantage that our approach can scale up to generate large training sets for many languages and domains. We experiment with several language pairs and data sources, and develop a variety of data filtering techniques. In the process, we explore how neural machine translation output differs from humanwritten sentences, finding clear differences in length, the amount of repetition, and the use of rare words. 1 | Learning Paraphrastic Sentence Embeddings from Back-Translated Bitext |
d44061090 | This paper describes the submissions to the efficiency track for GPUs at the Workshop for Neural Machine Translation and Generation by members of the University of Edinburgh, Adam Mickiewicz University, Tilde and University of Alicante. We focus on efficient implementation of the recurrent deep-learning model as implemented in Amun, the fast inference engine for neural machine translation. We improve the performance with an efficient mini-batching algorithm, and by fusing the softmax operation with the k-best extraction algorithm. Submissions using Amun were first, second and third fastest in the GPU efficiency track. | Fast Neural Machine Translation Implementation |
d202781916 | Ancient History relies on disciplines such as Epigraphy, the study of ancient inscribed texts, for evidence of the recorded past. However, these texts, "inscriptions", are often damaged over the centuries, and illegible parts of the text must be restored by specialists, known as epigraphists. This work presents PYTHIA, the first ancient text restoration model that recovers missing characters from a damaged text input using deep neural networks. Its architecture is carefully designed to handle longterm context information, and deal efficiently with missing or corrupted character and word representations. To train it, we wrote a nontrivial pipeline to convert PHI, the largest digital corpus of ancient Greek inscriptions, to machine actionable text, which we call PHI-ML. On PHI-ML, PYTHIA's predictions achieve a 30.1% character error rate, compared to the 57.3% of human epigraphists. Moreover, in 73.5% of cases the ground-truth sequence was among the Top-20 hypotheses of PYTHIA, which effectively demonstrates the impact of this assistive method on the field of digital epigraphy, and sets the state-of-the-art in ancient text restoration. | Restoring ancient text using deep learning: a case study on Greek epigraphy |
d258378280 | In the task of incremental few-shot relation classification, model performance is always limited by the incompatibility between the base feature embedding space and the novel feature embedding space. To tackle the issue, we propose a novel model named ICA-Proto: Iterative Cross Alignment prototypical network. Specifically, we incorporate the query representation into the encoding of novel prototypes and utilize the query-aware prototypes to update the query representation at the same time. Further, we implement the above process iteratively to achieve more interaction. In addition, a novel prototype quadruplet loss is designed to regulate the spatial distributions of embedding space, so as to make it easier for the relation classification. Experimental results on two benchmark datasets demonstrate that ICA-Proto significantly outperforms the state-of-the-art baseline model. | ICA-Proto: Iterative Cross Alignment Prototypical Network for Incremental Few-Shot Relation Classification |
d258378355 | Human conversations contain natural and reasonable topic shifts, reflected as the concept flows across utterances. Previous researches prove that explicitly modeling concept flows with a large commonsense knowledge graph effectively improves response quality. However, we argue that there exists a gap between the knowledge graph and the conversation. The knowledge graph has limited commonsense knowledge and ignores the characteristics of natural conversations. Thus, many concepts and relations in conversations are not included. To bridge this gap, we propose to enhance dialogue generation with conversational concept flows. Specifically, we extract abundant concepts and relations from natural conversations and build a new conversation-aware knowledge graph. In addition, we design a novel relationaware graph encoder to capture the concept flows guided by the knowledge graph. Experimental results on the large-scale Reddit conversation dataset indicate that our method performs better than strong baselines, and further analysis verifies the effectiveness of each com- | Enhancing Dialogue Generation with Conversational Concept Flows |
d226245980 | The extraction of interactions between chemicals and proteins from several biomedical articles is important in many fields of biomedical research such as drug development and prediction of drug side effects. Several natural language processing methods, including deep neural network (DNN) models, have been applied to address this problem. However, these methods were trained with hard-labeled data, which tend to become over-confident, leading to degradation of the model reliability.To estimate the data uncertainty and improve the reliability, "calibration" techniques have been applied to deep learning models. In this study, to extract chemical-protein interactions, we propose a DNN-based approach incorporating uncertainty information and calibration techniques. Our model first encodes the input sequence using a pre-trained languageunderstanding model, following which it is trained using two calibration methods: mixup training and addition of a confidence penalty loss. Finally, the model is re-trained with augmented data that are extracted using the estimated uncertainties. Our approach has achieved state-of-the-art performance with regard to the Biocreative VI ChemProt task, while preserving higher calibration abilities than those of previous approaches. Furthermore, our approach also presents the possibilities of using uncertainty estimation for performance improvement. | Extracting Chemical-Protein Interactions via Calibrated Deep Neural Network and Self-training |
d234341857 | In human-level NLP tasks, such as predicting mental health, personality, or demographics, the number of observations is often smaller than the standard 768+ hidden state sizes of each layer within modern transformer-based language models, limiting the ability to effectively leverage transformers. Here, we provide a systematic study on the role of dimension reduction methods (principal components analysis, factorization techniques, or multi-layer auto-encoders) as well as the dimensionality of embedding vectors and sample sizes as a function of predictive performance. We first find that fine-tuning large models with a limited amount of data pose a significant difficulty which can be overcome with a pre-trained dimension reduction regime. RoBERTa consistently achieves top performance in humanlevel tasks, with PCA giving benefit over other reduction methods in better handling users that write longer texts. Finally, we observe that a majority of the tasks achieve results comparable to the best performance with just 1 12 of the embedding dimensions. | Empirical Evaluation of Pre-trained Transformers for Human-Level NLP: The Role of Sample Size and Dimensionality |
d207870601 | We analyse coreference phenomena in three neural machine translation systems trained with different data settings with or without access to explicit intra-and cross-sentential anaphoric information. We compare system performance on two different genres: news and TED talks. To do this, we manually annotate (the possibly incorrect) coreference chains in the MT outputs and evaluate the coreference chain translations. We define an error typology that aims to go further than pronoun translation adequacy and includes types such as incorrect word selection or missing words. The features of coreference chains in automatic translations are also compared to those of the source texts and human translations. The analysis shows stronger potential translationese effects in machine translated outputs than in human translations. | Analysing Coreference in Transformer Outputs |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.