text stringlengths 82 736 | label int64 0 1 |
|---|---|
to build a baseline smt system , we used the stanford phrasal library trained on europarl corpus---we used the sentence-aligned europarl corpus for the construction of our wsd module | 1 |
we have presented a framework to refine non-terminal math-w-17-1-0-9 in hierarchical translation rules with semantic representations---we propose a framework to refine nonterminals in hierarchical translation rules with real-valued semantic representations | 1 |
however , compared to the baselines , the contribution of syntactic structure is not significant to the overall performance---to us , our proposed model uses an endto-end trainable neural architecture to predict commentaries given the game state | 0 |
furthermore , we train a 5-gram language model using the sri language toolkit---we use srilm for training a trigram language model on the english side of the training corpus | 1 |
we trained a 4-gram language model on this data with kneser-ney discounting using srilm---we trained a trigram language model on the chinese side , with the srilm toolkit , using the modified kneser-ney smoothing option | 1 |
note that math-w-3-1-1-52 , the empty string---math-w-17-1-1-46 ( the context around math-w-17-1-1-55 ) | 1 |
the probabilistic language model is constructed on google web 1t 5-gram corpus by using the srilm toolkit---a 4-gram language model was trained on the target side of the parallel data using the srilm toolkit from stolcke | 1 |
the grosz and sidner model has a much broader scope---one and the more sophisticated model proposed by grosz and sidner | 1 |
we used the srilm toolkit to train a 4-gram language model on the english side of the training corpus---we used a 5-gram language model with modified kneser-ney smoothing , built with the srilm toolkit | 1 |
we also used a projective english dataset derived from the penn treebank by applying the standard head rules of yamada and matsumoto---we use the wsj portion of the penn treebank 4 , augmented with head-dependant information using the rules of yamada and matsumoto | 1 |
furthermore , we propose to utilize the sentence-level bleu as the specific objective for the generator---turney uses the number of hits returned by a web search engine to calculate the pointwise mutual information between terms , as an indicator of synonymy | 0 |
badjatiya et al presented a gradient boosted lstm model with random embeddings to outperform state of the art hate speech detection techniques---among several prominent works , badjatiya et al employed multiple deep learning architectures including cnns , lstms , and fasttext to learn semantic word embeddings for hate speech detection | 1 |
mcinnes et al made use of concept unique identifiers from umls which are also assigned by metamap---we follow the approach presented by mcinnes et al to generate features based on umls concept unique identifiers | 1 |
our results show that tweets do contain signals indicative of purchase stages---our results indicate that tweets indeed contain signals indicative of purchase stages | 1 |
this paper reports on the first step aimed at automatically assigning importance scores to parts of the lecture---experimental results show that our system outperforms the base system with a 3 . 4 % gain in f1 , and generates logical forms more accurately | 0 |
for example , data-driven approaches for discovering dialogue structure have been applied to corpora of human-human task-oriented dialogue using general models of task structure---in task-oriented domains , recent work has approached dialogue act classification by learning dialogue management models entirely from human-human corpora | 1 |
hochreiter and schmidhuber , 1997 ) proposed a long short-term memory network , which can be used for sequence processing tasks---hochreiter and schmidhuber proposed long short-term memories as the specific version of rnn designed to overcome vanishing and exploding gradient problem | 1 |
knowledge graphs such as freebase , yago and wordnet are among the most widely used resources in nlp applications---hand-built lexicons , such as cyc and wordnet , are the most useful to provide resources for nlp applications | 1 |
this algorithm approximates the data within a specified memory bound while preserving the covariance structure necessary for pca---algorithm approximates the data within a specified memory bound while preserving the covariance structure necessary for pca | 1 |
we propose a translation framework based on situation theory---in this paper , we propose a framework based on situation theory | 1 |
we are able to surpass human recall and achieve an f1 of 0.51 on a question-answering task with less than 50 hours of effort using a hybrid approach that mixes active learning , bootstrapping , and limited ( 5 hours ) manual rule writing---through machine learning and limited human pattern writing ( 6 hours ) , we adapted a machine reading system within a week ( using less than 50 person hours ) , achieving question answering performance with an f1 of 0 . 5 and with recall | 1 |
we tune the feature weights with batch k-best mira to maximize bleu on a development set---in order to tune all systems , we use the k-best batch mira | 1 |
the automatic classification results were compared with the manual judgement of several linguistics students---automatic classification results were compared with a baseline method and with the manual judgement of several linguistics students | 1 |
we combine the extractive model and the abstractive model---we propose a guiding generation model for abstractive text | 1 |
by designing a two player game , we can both collect and verify referring expressions directly within the game---and all instances in both languages are then fed into a bilingual active learning engine | 0 |
combinatory categorial grammars are a linguistically-motivated model for a wide range of language phenomena---ccg is a linguistically-motivated categorial formalism for modeling a wide range of language phenomena | 1 |
we further used a 5-gram language model trained using the srilm toolkit with modified kneser-ney smoothing---we use a fourgram language model with modified kneser-ney smoothing as implemented in the srilm toolkit | 1 |
statistical significance of system differences in terms of f1 was assessed by an approximate randomization test---we assessed the statistical significance of f-measure improvements over baseline , using the approximate randomness test | 1 |
we can use an automatic evaluation measure such as bleu as ev---following wan et al , we use the bleu metric for string comparison | 1 |
semantic parsing is the problem of mapping natural language strings into meaning representations---we demonstrate that concept drift is a real , pervasive issue for learning from issue | 0 |
in this paper , we develop novel techniques to characterize the behavior of vqa models---in this paper we propose systematic methods to analyze the behavior of these models | 1 |
we further used a 5-gram language model trained using the srilm toolkit with modified kneser-ney smoothing---the language models in this experiment were trigram models with good-turing smoothing built using srilm | 1 |
we present a representation for common sense spatial knowledge and an approach to extract it from 3d scene data---we present a spatial knowledge representation that can be learned from 3d scenes | 1 |
this means in practice that the language model was trained using the srilm toolkit---these language models were built up to an order of 5 with kneser-ney smoothing using the srilm toolkit | 1 |
bollen et al used the mood dimension , calm together with the index value itself to predict the dow jones industrial average---for two grammars in different languages , an adequate algorithm should both improve the cross-lingual similarity between two grammars and maintain the non-triviality of each grammar , where non-triviality | 0 |
recurrent neural networks are a type of neural networks in which the hidden layer is connected to itself so that the previous hidden state is used along with the input at the current step---word sense disambiguation ( wsd ) is the task of determining the meaning of a word in a given context | 0 |
for both languages , we used the srilm toolkit to train a 5-gram language model using all monolingual data provided---we use our system combination module which includes a language modeling tool , a mert process , and mbr decoding of its own | 0 |
in this work , we employ the toolkit word2vec to pre-train the word embedding for the source and target languages---we use the penn tree bank , constructed from articles from the wall street journal , as our primary training corpus , with the standard training split of 42068 sentences | 0 |
in the future , we plan to optimize feature weights for max-translation decoding directly on the entire packed translation hypergraph rather than on n-best derivations , following the lattice-based mert ( cite-p-18-1-15 )---for finding optimal translations , we extend the minimum error rate training ( mert ) algorithm ( cite-p-18-1-21 ) to tune feature weights with respect to bleu score for max-translation decoding | 1 |
in this study , we propose an innovative sentence compression model based on expanded constituent parse trees---in this paper , we focus on the problem of using sentence compression techniques | 1 |
huang et al improve a bigram hmm pos tagger by latent annotation and self-training---semantic parsing is the task of translating natural language utterances into a machine-interpretable meaning representation | 0 |
in this paper , an efficient disfluency detection approach based on right-to-left transition-based parsing is proposed , which can efficiently identify disfluencies and keep asr outputs grammatical---in this paper , we propose detecting disfluencies using a right-to-left transition-based dependency parsing ( r2l parsing ) , where the words are consumed from right to left to build the parsing tree | 1 |
the optimisation of the feature weights of the model is done with minimum error rate training against the bleu evaluation metric---dependency parsing is a fundamental task for language processing which has been investigated for decades | 0 |
word sense disambiguation ( wsd ) is the task of determining the meaning of a word in a given context---word sense disambiguation ( wsd ) is the task of automatically determining the correct sense for a target word given the context in which it occurs | 1 |
the importance of robust techniques for predicate-argument transformation has motivated the development of large-scale text corpora with predicate-argument annotations such as propbank and framenet---recently , the integration of nlp systems with manually-built resources at the predicate argument-level , such as framenet and propbank has received growing interest | 1 |
we propose a method to learn bilingual word embeddings as the input of cnn , using only a bilingual dictionary and unlabeled corpus---in this paper , we propose an efficient way to model cross-lingual sentences with only a bilingual dictionary and dependency | 1 |
their word embeddings were generated with word2vec , and trained on the arabic gigaword corpus---the neural embeddings were created using the word2vec software 3 accompanying | 1 |
for both languages , we used the srilm toolkit to train a 5-gram language model using all monolingual data provided---we trained a specific language model using srilm from each of these corpora in order to estimate n-gram log-probabilities | 1 |
to this end , our goal is to incorporate the gloss information into a unified neural network for all of the polysemous words---in this paper , we integrate the context and glosses of the target word into a unified framework | 1 |
paraphrases can be viewed as bidirectional entailment rules---rules are not paraphrases but rather one-directional entailment rules | 1 |
in this paper , we study the utility of the discourse structure on the user side of a dialogue system---we propose using a graphical representation of the discourse structure as a way of improving the performance of complex-domain dialogue systems | 1 |
following previous work , we believe that using sequential information rather than a bag-of-words model would help improve performance---we follow a previous attempt to use a sequence-to-sequence learning model augmented with the attention mechanism | 1 |
while the former perform best in isolation , the latter present a scalable alternative within joint systems---the translation quality is evaluated by case-insensitive bleu and ter metric | 0 |
research into using distributional information in srl dates back to gildea and jurafsky , who used distributions over verb-object co-occurrence clusters to improve coverage in argument classification---early work in srl dates back to gildea and jurafsky , who were the first to model role assignment to verb arguments based on framenet | 1 |
the comparison with voice recognition and a screen keyboard showed koosho can be a more practical solution compared to the screen keyboard---the comparison with voice recognition and a screen keyboard showed that koosho can be a more practical solution | 1 |
peldszus and stede used decoding based on minimum spanning trees to jointly predict argumentative segments and their types as well as argumentative relations , to generate an argumentation graph from text---peldszus and stede demonstrate how the resulting graphs of argument components and their relations can be parsed into discourse structure | 1 |
topic modeling is an unsupervised method to cluster documents based on context information---topic modelling is a popular statistical method for clustering documents | 1 |
we use the word and context vectors released by which was shown to perform strongly on lexical substitution task---we use the word and context vectors released by melamud et al , 5 which were previously shown to perform strongly in lexical substitution tasks | 1 |
the feature weights were tuned on the wmt newstest2008 development set using mert---the smt weighting parameters were tuned by mert in the development data | 1 |
for training our system classifier , we have used scikit-learn---we use the skll and scikit-learn toolkits | 1 |
some of the recent works that have employed pre-trained language models include ulmfit , elmo , glomo , bert and openai transformer---we use mini-batch update and adagrad to optimize the parameter learning | 0 |
this paper presents woe , an open ie system which improves dramatically on textrunner ’ s precision and recall---this paper introduces woe , a new approach to open ie that uses self-supervised learning | 1 |
the world wide web ( www ) is the largest repository of information known to mankind---world wide web ( www ) is the most useful and powerful information dissemination system on the internet | 1 |
the language modelling approach proved very effective for the information retrieval task---this approach showed improvements over the baseline language modelling approach | 1 |
we used a script from with 89 , 500 merge operations---both planas and furuse and hodasz and pohl proposed to use lemma and parts of speech along with surface form comparison | 0 |
to set the weights , 位 m , we performed minimum error rate training on the development set using bleu as the objective function---however , this is cumbersome and time-consuming for large corpora | 0 |
finally , in section 5 we draw the conclusions of our study and describe our plans for extending the method---kawahara and uchimoto used a separately trained binary classifier to select sentences as additional training data | 0 |
we use 300-dimensional word embeddings from glove to initialize the model---we use pre-trained glove vector for initialization of word embeddings | 1 |
next , tens of words are sampled from each group---next , a fixed number of words are randomly sampled from each group | 1 |
we also use a 4-gram language model trained using srilm with kneser-ney smoothing---for probabilities , we trained 5-gram language models using srilm | 1 |
we are interested in collecting unknown words which occur more than once throughout the corpus---in this paper , we are interested in extracting the unknown words with high precision and recall | 1 |
in contrast , supervised path-based methods can capture relational information between two words---although path-based methods can capture the relational information between two words | 1 |
semantic parsing is the task of mapping a natural language ( nl ) sentence into a completely formal meaning representation ( mr ) or logical form---semantic parsing is the task of mapping natural language sentences to a formal representation of meaning | 1 |
experiments showed that bilingual co-training is effective for improving the performance of classifiers in both languages---similar to bilingual co-training , classifiers for two languages cooperated in learning with bilingual resources | 1 |
rhetorical structure theory is a well known text representation technique that represents the knowledge present in the text using semantic relations known as discourse relations---rhetorical structure theory is a framework for describing the organization of a text and what a text conveys by identifying hierarchical structures in text | 1 |
the system mostly follows the standard encoder-decoder architecture using rnn layers and attention mechanism---we used the target side of the parallel corpus and the srilm toolkit to train a 5-gram language model | 0 |
in this paper , we present a comprehensive exploration of syntactic elements in writing styles , with particular emphasis on interpretable characterization of stylistic elements---in this paper , we have presented a comprehensive exploration of syntactic elements in writing styles , with particular emphasis on interpretable characterization of stylistic elements | 1 |
as a baseline system , we used the moses statistical machine translation package to build grapheme-based and phoneme-based translation systems , using a bigram language model---we used the open source moses phrase-based mt system to test the impact of the preprocessing technique on translation results | 1 |
wordnet is a key lexical resource for natural language applications---the 5-gram target language model was trained using kenlm | 0 |
in conclusion , we have shown how to create long-span aac language models using openly available resources---we use glove vectors with 100 dimensions trained on wikipedia and gigaword as word embeddings , which we do not optimize during training | 0 |
in philosophy and linguistics , it is generally accepted that negation conveys positive meaning---in philosophy and linguistics , it is generally accepted that negation conveys positive meanings | 1 |
sarcasm is a form of verbal irony that is intended to express contempt or ridicule---sarcasm is a form of speech in which speakers say the opposite of what they truly mean in order to convey a strong sentiment | 1 |
this is therefore the underlying approach for reducing the word sampling problem into graph-based active learning---by showing that the existing heuristic sampling approach is simply a special case of a graph-based active learning | 1 |
in this paper we start from a trained word embedding space , and learn a manifold from it to improve results---in this paper , we re-embed pre-trained word embeddings with a stage of manifold learning | 1 |
snyder and barzilay propose a discriminative model for unsupervised morphological segmentation by using morphological chains to model the word formation process---snyder and barzilay extend the approach to unsupervised annotation of morphology in semitic languages via a hierarchical bayesian network | 1 |
collobert and weston propose a unified deep convolutional neural network for different tasks by using a set of taskindependent word embeddings together with a set of task-specific word embeddings---we used the srilm toolkit and kneser-ney discounting for estimating 5-grams lms | 0 |
here we use the most widely used long short term memory network as our composition model---therefore , we use the long short-term memory network to overcome this problem | 1 |
we derive 100-dimensional word vectors using word2vec skip-gram model trained over the domain corpus---we investigate a novel method to detect asymmetric entailment | 0 |
word alignment is the task of identifying translational relations between words in parallel corpora , in which a word at one language is usually translated into several words at the other language ( fertility model ) ( cite-p-18-1-0 )---entity resolution ( er ) is the process of associating mentions of entities in text with a dictionary of entities | 0 |
jackendoff and others have proposed that lexical rules be interpreted as redundancy statements that abbreviate the statement of the lexicon but that are not applied generatively---jackendoff and others have proposed that lexical rules be interpreted as redundancy statements which abbreviate the statement of the lexicon but which are not applied generatively | 1 |
coreference resolution is a task aimed at identifying phrases ( mentions ) referring to the same entity---since coreference resolution is a pervasive discourse phenomenon causing performance impediments in current ie systems , we considered a corpus of aligned english and romanian texts to identify coreferring expressions | 1 |
we also obtain the embeddings of each word from word2vec---we use the pre-trained word2vec embeddings provided by mikolov et al as model input | 1 |
we use srilm for training a trigram language model on the english side of the training corpus---we use srilm for training the 5-gram language model with interpolated modified kneser-ney discounting | 1 |
we used a 5-gram language model with modified kneser-ney smoothing implemented using the srilm toolkit---of this corpus , global inference is applied to provide more confident and informative data | 0 |
nakov and hearst demonstrate that web counts can aid in identifying the bracketing in higher-arity noun compounds---nakov and hearst demonstrated the effectiveness of using search engine statistics to improve the noun compound bracketing | 1 |
coreference resolution is a challenging task , that involves identification and clustering of noun phrases mentions that refer to the same real-world entity---this probability is defined as the possibly infinite sum of the probabilities of all strings of the form xwy , for any pair of strings math-w-2-3-1-89 and math-w-2-3-1-91 over the alphabet of math-w-2-3-1-96 | 0 |
coreference resolution is a well known clustering task in natural language processing---coreference resolution is a set partitioning problem in which each resulting partition refers to an entity | 1 |
a gaussian process is a stochastic process that defines a nonparametric prior over functions in bayesian statistics---a gaussian process is a generative model of bayesian inference that can be used for function regression | 1 |
sentence compression is the task of compressing long , verbose sentences into short , concise ones---sentence compression is a complex paraphrasing task with information loss involving substitution , deletion , insertion , and reordering operations | 1 |
the language model is a 5-gram with interpolation and kneser-ney smoothing---the language models were trained with kneser-ney backoff smoothing using the sri language modeling toolkit , | 1 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.