text
stringlengths
49
577
label
stringclasses
7 values
metadata
sequence
The agreement in question involves number in [[ nouns ]] and << reflexive pronouns >> and is syntactic rather than semantic in nature because grammatical number in English , like grammatical gender in languages such as French , is partly arbitrary .
CONJUNCTION
[ 7, 7, 9, 10 ]
The agreement in question involves number in nouns and reflexive pronouns and is syntactic rather than semantic in nature because grammatical number in English , like [[ grammatical gender ]] in << languages >> such as French , is partly arbitrary .
FEATURE-OF
[ 26, 27, 29, 29 ]
The agreement in question involves number in nouns and reflexive pronouns and is syntactic rather than semantic in nature because grammatical number in English , like grammatical gender in << languages >> such as [[ French ]] , is partly arbitrary .
HYPONYM-OF
[ 32, 32, 29, 29 ]
In this paper , a novel [[ method ]] to learn the << intrinsic object structure >> for robust visual tracking is proposed .
USED-FOR
[ 6, 6, 10, 12 ]
In this paper , a novel method to learn the [[ intrinsic object structure ]] for << robust visual tracking >> is proposed .
USED-FOR
[ 10, 12, 14, 16 ]
The basic assumption is that the << parameterized object state >> lies on a [[ low dimensional manifold ]] and can be learned from training data .
FEATURE-OF
[ 12, 14, 6, 8 ]
Based on this assumption , firstly we derived the [[ dimensionality reduction and density estimation algorithm ]] for << unsupervised learning of object intrinsic representation >> , the obtained non-rigid part of object state reduces even to 2 dimensions .
USED-FOR
[ 9, 14, 16, 21 ]
Secondly the << dynamical model >> is derived and trained based on this [[ intrinsic representation ]] .
USED-FOR
[ 11, 12, 2, 3 ]
Thirdly the learned [[ intrinsic object structure ]] is integrated into a << particle-filter style tracker >> .
PART-OF
[ 3, 5, 10, 12 ]
We will show that this intrinsic object representation has some interesting properties and based on which the newly derived [[ dynamical model ]] makes << particle-filter style tracker >> more robust and reliable .
USED-FOR
[ 19, 20, 22, 24 ]
Experiments show that the learned [[ tracker ]] performs much better than existing << trackers >> on the tracking of complex non-rigid motions such as fish twisting with self-occlusion and large inter-frame lip motion .
COMPARE
[ 5, 5, 11, 11 ]
Experiments show that the learned [[ tracker ]] performs much better than existing trackers on the << tracking of complex non-rigid motions >> such as fish twisting with self-occlusion and large inter-frame lip motion .
USED-FOR
[ 5, 5, 14, 18 ]
Experiments show that the learned tracker performs much better than existing [[ trackers ]] on the << tracking of complex non-rigid motions >> such as fish twisting with self-occlusion and large inter-frame lip motion .
USED-FOR
[ 11, 11, 14, 18 ]
Experiments show that the learned tracker performs much better than existing trackers on the tracking of << complex non-rigid motions >> such as [[ fish twisting ]] with self-occlusion and large inter-frame lip motion .
HYPONYM-OF
[ 21, 22, 16, 18 ]
Experiments show that the learned tracker performs much better than existing trackers on the tracking of complex non-rigid motions such as << fish twisting >> with [[ self-occlusion ]] and large inter-frame lip motion .
FEATURE-OF
[ 24, 24, 21, 22 ]
Experiments show that the learned tracker performs much better than existing trackers on the tracking of complex non-rigid motions such as fish twisting with [[ self-occlusion ]] and large << inter-frame lip motion >> .
CONJUNCTION
[ 24, 24, 27, 29 ]
Experiments show that the learned tracker performs much better than existing trackers on the tracking of complex non-rigid motions such as << fish twisting >> with self-occlusion and large [[ inter-frame lip motion ]] .
FEATURE-OF
[ 27, 29, 21, 22 ]
The proposed [[ method ]] also has the potential to solve other type of << tracking problems >> .
USED-FOR
[ 2, 2, 12, 13 ]
In this paper , we present a [[ digital signal processor -LRB- DSP -RRB- implementation ]] of << real-time statistical voice conversion -LRB- VC -RRB- >> for silent speech enhancement and electrolaryngeal speech enhancement .
USED-FOR
[ 7, 13, 15, 21 ]
In this paper , we present a digital signal processor -LRB- DSP -RRB- implementation of [[ real-time statistical voice conversion -LRB- VC -RRB- ]] for << silent speech enhancement >> and electrolaryngeal speech enhancement .
USED-FOR
[ 15, 21, 23, 25 ]
In this paper , we present a digital signal processor -LRB- DSP -RRB- implementation of [[ real-time statistical voice conversion -LRB- VC -RRB- ]] for silent speech enhancement and << electrolaryngeal speech enhancement >> .
USED-FOR
[ 15, 21, 27, 29 ]
In this paper , we present a digital signal processor -LRB- DSP -RRB- implementation of real-time statistical voice conversion -LRB- VC -RRB- for [[ silent speech enhancement ]] and << electrolaryngeal speech enhancement >> .
CONJUNCTION
[ 23, 25, 27, 29 ]
[[ Electrolaryngeal speech ]] is one of the typical types of << alaryngeal speech >> produced by an alternative speaking method for laryngectomees .
HYPONYM-OF
[ 0, 1, 9, 10 ]
Electrolaryngeal speech is one of the typical types of << alaryngeal speech >> produced by an alternative [[ speaking method ]] for laryngectomees .
USED-FOR
[ 15, 16, 9, 10 ]
Electrolaryngeal speech is one of the typical types of alaryngeal speech produced by an alternative [[ speaking method ]] for << laryngectomees >> .
USED-FOR
[ 15, 16, 18, 18 ]
However , the [[ sound quality ]] of << NAM and electrolaryngeal speech >> suffers from lack of naturalness .
EVALUATE-FOR
[ 3, 4, 6, 9 ]
VC has proven to be one of the promising approaches to address this problem , and << it >> has been successfully implemented on [[ devices ]] with sufficient computational resources .
USED-FOR
[ 22, 22, 16, 16 ]
VC has proven to be one of the promising approaches to address this problem , and it has been successfully implemented on << devices >> with [[ sufficient computational resources ]] .
FEATURE-OF
[ 24, 26, 22, 22 ]
An implementation on << devices >> that are highly portable but have [[ limited computational resources ]] would greatly contribute to its practical use .
FEATURE-OF
[ 10, 12, 3, 3 ]
In this paper we further implement << real-time VC >> on a [[ DSP ]] .
USED-FOR
[ 10, 10, 6, 7 ]
To implement the two << speech enhancement systems >> based on [[ real-time VC ]] , one from NAM to a whispered voice and the other from electrolaryngeal speech to a natural voice , we propose several methods for reducing computational cost while preserving conversion accuracy .
USED-FOR
[ 9, 10, 4, 6 ]
To implement the two << speech enhancement systems >> based on real-time VC , [[ one ]] from NAM to a whispered voice and the other from electrolaryngeal speech to a natural voice , we propose several methods for reducing computational cost while preserving conversion accuracy .
HYPONYM-OF
[ 12, 12, 4, 6 ]
To implement the two speech enhancement systems based on real-time VC , [[ one ]] from NAM to a whispered voice and the << other >> from electrolaryngeal speech to a natural voice , we propose several methods for reducing computational cost while preserving conversion accuracy .
CONJUNCTION
[ 12, 12, 21, 21 ]
To implement the two << speech enhancement systems >> based on real-time VC , one from NAM to a whispered voice and the [[ other ]] from electrolaryngeal speech to a natural voice , we propose several methods for reducing computational cost while preserving conversion accuracy .
HYPONYM-OF
[ 21, 21, 4, 6 ]
To implement the two speech enhancement systems based on real-time VC , one from NAM to a whispered voice and the other from electrolaryngeal speech to a natural voice , we propose several << methods >> for reducing [[ computational cost ]] while preserving conversion accuracy .
EVALUATE-FOR
[ 36, 37, 33, 33 ]
To implement the two speech enhancement systems based on real-time VC , one from NAM to a whispered voice and the other from electrolaryngeal speech to a natural voice , we propose several methods for reducing [[ computational cost ]] while preserving << conversion accuracy >> .
CONJUNCTION
[ 36, 37, 40, 41 ]
To implement the two speech enhancement systems based on real-time VC , one from NAM to a whispered voice and the other from electrolaryngeal speech to a natural voice , we propose several << methods >> for reducing computational cost while preserving [[ conversion accuracy ]] .
EVALUATE-FOR
[ 40, 41, 33, 33 ]
We conduct experimental evaluations and show that << real-time VC >> is capable of running on a [[ DSP ]] with little degradation .
USED-FOR
[ 15, 15, 7, 8 ]
We propose a [[ method ]] that automatically generates << paraphrase >> sets from seed sentences to be used as reference sets in objective machine translation evaluation measures like BLEU and NIST .
USED-FOR
[ 3, 3, 7, 7 ]
We propose a method that automatically generates [[ paraphrase ]] sets from seed sentences to be used as reference sets in objective << machine translation evaluation measures >> like BLEU and NIST .
USED-FOR
[ 7, 7, 20, 23 ]
We propose a method that automatically generates paraphrase sets from seed sentences to be used as reference sets in objective << machine translation evaluation measures >> like [[ BLEU ]] and NIST .
HYPONYM-OF
[ 25, 25, 20, 23 ]
We propose a method that automatically generates paraphrase sets from seed sentences to be used as reference sets in objective machine translation evaluation measures like [[ BLEU ]] and << NIST >> .
CONJUNCTION
[ 25, 25, 27, 27 ]
We propose a method that automatically generates paraphrase sets from seed sentences to be used as reference sets in objective << machine translation evaluation measures >> like BLEU and [[ NIST ]] .
HYPONYM-OF
[ 27, 27, 20, 23 ]
We measured the quality of the paraphrases produced in an experiment , i.e. , -LRB- i -RRB- their << grammaticality >> : at least 99 % correct sentences ; -LRB- ii -RRB- their [[ equivalence in meaning ]] : at least 96 % correct paraphrases either by meaning equivalence or entailment ; and , -LRB- iii -RRB- the amount of internal lexical and syntactical variation in a set of paraphrases : slightly superior to that of hand-produced sets .
CONJUNCTION
[ 31, 33, 18, 18 ]
We measured the quality of the paraphrases produced in an experiment , i.e. , -LRB- i -RRB- their grammaticality : at least 99 % correct sentences ; -LRB- ii -RRB- their equivalence in meaning : at least 96 % correct << paraphrases >> either by [[ meaning equivalence ]] or entailment ; and , -LRB- iii -RRB- the amount of internal lexical and syntactical variation in a set of paraphrases : slightly superior to that of hand-produced sets .
USED-FOR
[ 43, 44, 40, 40 ]
We measured the quality of the paraphrases produced in an experiment , i.e. , -LRB- i -RRB- their grammaticality : at least 99 % correct sentences ; -LRB- ii -RRB- their equivalence in meaning : at least 96 % correct paraphrases either by [[ meaning equivalence ]] or << entailment >> ; and , -LRB- iii -RRB- the amount of internal lexical and syntactical variation in a set of paraphrases : slightly superior to that of hand-produced sets .
CONJUNCTION
[ 43, 44, 46, 46 ]
We measured the quality of the paraphrases produced in an experiment , i.e. , -LRB- i -RRB- their grammaticality : at least 99 % correct sentences ; -LRB- ii -RRB- their equivalence in meaning : at least 96 % correct << paraphrases >> either by meaning equivalence or [[ entailment ]] ; and , -LRB- iii -RRB- the amount of internal lexical and syntactical variation in a set of paraphrases : slightly superior to that of hand-produced sets .
USED-FOR
[ 46, 46, 40, 40 ]
We measured the quality of the paraphrases produced in an experiment , i.e. , -LRB- i -RRB- their grammaticality : at least 99 % correct sentences ; -LRB- ii -RRB- their << equivalence in meaning >> : at least 96 % correct paraphrases either by meaning equivalence or entailment ; and , -LRB- iii -RRB- the amount of [[ internal lexical and syntactical variation ]] in a set of paraphrases : slightly superior to that of hand-produced sets .
CONJUNCTION
[ 56, 60, 31, 33 ]
We measured the quality of the paraphrases produced in an experiment , i.e. , -LRB- i -RRB- their grammaticality : at least 99 % correct sentences ; -LRB- ii -RRB- their equivalence in meaning : at least 96 % correct paraphrases either by meaning equivalence or entailment ; and , -LRB- iii -RRB- the amount of internal lexical and syntactical variation in a set of [[ paraphrases ]] : slightly superior to that of << hand-produced sets >> .
COMPARE
[ 65, 65, 72, 73 ]
The << paraphrase >> sets produced by this [[ method ]] thus seem adequate as reference sets to be used for MT evaluation .
USED-FOR
[ 6, 6, 1, 1 ]
[[ Graph unification ]] remains the most expensive part of << unification-based grammar parsing >> .
PART-OF
[ 0, 1, 8, 10 ]
We focus on one [[ speed-up element ]] in the design of << unification algorithms >> : avoidance of copying of unmodified subgraphs .
PART-OF
[ 4, 5, 10, 11 ]
We propose a << method >> of attaining such a design through a method of [[ structure-sharing ]] which avoids log -LRB- d -RRB- overheads often associated with structure-sharing of graphs without any use of costly dependency pointers .
USED-FOR
[ 13, 13, 3, 3 ]
The proposed [[ scheme ]] eliminates redundant copying while maintaining the quasi-destructive scheme 's ability to avoid over copying and early copying combined with its ability to handle << cyclic structures >> without algorithmic additions .
USED-FOR
[ 2, 2, 26, 27 ]
The proposed << scheme >> eliminates redundant copying while maintaining the [[ quasi-destructive scheme 's ability ]] to avoid over copying and early copying combined with its ability to handle cyclic structures without algorithmic additions .
FEATURE-OF
[ 9, 12, 2, 2 ]
The proposed scheme eliminates redundant copying while maintaining the quasi-destructive scheme 's ability to avoid [[ over copying ]] and << early copying >> combined with its ability to handle cyclic structures without algorithmic additions .
CONJUNCTION
[ 15, 16, 18, 19 ]
We describe a novel technique and implemented [[ system ]] for constructing a << subcategorization dictionary >> from textual corpora .
USED-FOR
[ 7, 7, 11, 12 ]
We describe a novel technique and implemented << system >> for constructing a subcategorization dictionary from [[ textual corpora ]] .
USED-FOR
[ 14, 15, 7, 7 ]
We also demonstrate that a << subcategorization dictionary >> built with the [[ system ]] improves the accuracy of a parser by an appreciable amount
USED-FOR
[ 10, 10, 5, 6 ]
We also demonstrate that a subcategorization dictionary built with the system improves the [[ accuracy ]] of a << parser >> by an appreciable amount
EVALUATE-FOR
[ 13, 13, 16, 16 ]
We also demonstrate that a << subcategorization dictionary >> built with the system improves the accuracy of a [[ parser ]] by an appreciable amount
EVALUATE-FOR
[ 16, 16, 5, 6 ]
A number of powerful << registration criteria >> have been developed in the last decade , most prominently the criterion of [[ maximum mutual information ]] .
HYPONYM-OF
[ 19, 21, 4, 5 ]
Although this criterion provides for good registration results in many applications , << it >> remains a purely [[ low-level criterion ]] .
FEATURE-OF
[ 16, 17, 12, 12 ]
In this paper , we will develop a [[ Bayesian framework ]] that allows to impose statistically learned prior knowledge about the joint intensity distribution into << image registration methods >> .
USED-FOR
[ 8, 9, 24, 26 ]
In this paper , we will develop a Bayesian framework that allows to impose [[ statistically learned prior knowledge ]] about the joint intensity distribution into << image registration methods >> .
USED-FOR
[ 14, 17, 24, 26 ]
In this paper , we will develop a Bayesian framework that allows to impose << statistically learned prior knowledge >> about the [[ joint intensity distribution ]] into image registration methods .
FEATURE-OF
[ 20, 22, 14, 17 ]
The << prior >> is given by a [[ kernel density estimate ]] on the space of joint intensity distributions computed from a representative set of pre-registered image pairs .
USED-FOR
[ 6, 8, 1, 1 ]
The prior is given by a [[ kernel density estimate ]] on the space of << joint intensity distributions >> computed from a representative set of pre-registered image pairs .
USED-FOR
[ 6, 8, 13, 15 ]
The prior is given by a kernel density estimate on the space of << joint intensity distributions >> computed from a representative set of [[ pre-registered image pairs ]] .
USED-FOR
[ 22, 24, 13, 15 ]
Experimental results demonstrate that the resulting [[ registration process ]] is more robust to << missing low-level information >> as it favors intensity correspondences statistically consistent with the learned intensity distributions .
USED-FOR
[ 6, 7, 12, 14 ]
Experimental results demonstrate that the resulting registration process is more robust to missing low-level information as [[ it ]] favors << intensity correspondences >> statistically consistent with the learned intensity distributions .
USED-FOR
[ 16, 16, 18, 19 ]
We present a [[ method ]] for << synthesizing complex , photo-realistic facade images >> , from a single example .
USED-FOR
[ 3, 3, 5, 10 ]
After parsing the example image into its << semantic components >> , a [[ tiling ]] for it is generated .
USED-FOR
[ 11, 11, 7, 8 ]
Novel tilings can then be created , yielding << facade textures >> with different dimensions or with [[ occluded parts inpainted ]] .
FEATURE-OF
[ 15, 17, 8, 9 ]
A [[ genetic algorithm ]] guides the novel << facades >> as well as inpainted parts to be consistent with the example , both in terms of their overall structure and their detailed textures .
USED-FOR
[ 1, 2, 6, 6 ]
A [[ genetic algorithm ]] guides the novel facades as well as << inpainted parts >> to be consistent with the example , both in terms of their overall structure and their detailed textures .
USED-FOR
[ 1, 2, 10, 11 ]
Promising results for [[ multiple standard datasets ]] -- in particular for the different building styles they contain -- demonstrate the potential of the << method >> .
EVALUATE-FOR
[ 3, 5, 22, 22 ]
We introduce a new << interactive corpus exploration tool >> called [[ InfoMagnets ]] .
HYPONYM-OF
[ 9, 9, 4, 7 ]
[[ InfoMagnets ]] aims at making << exploratory corpus analysis >> accessible to researchers who are not experts in text mining .
USED-FOR
[ 0, 0, 4, 6 ]
As evidence of its usefulness and usability , [[ it ]] has been used successfully in a research context to uncover relationships between language and behavioral patterns in two distinct << domains >> : tutorial dialogue -LRB- Kumar et al. , submitted -RRB- and on-line communities -LRB- Arguello et al. , 2006 -RRB- .
USED-FOR
[ 8, 8, 28, 28 ]
As evidence of its usefulness and usability , it has been used successfully in a research context to uncover relationships between language and behavioral patterns in two distinct << domains >> : [[ tutorial dialogue ]] -LRB- Kumar et al. , submitted -RRB- and on-line communities -LRB- Arguello et al. , 2006 -RRB- .
HYPONYM-OF
[ 30, 31, 28, 28 ]
As evidence of its usefulness and usability , it has been used successfully in a research context to uncover relationships between language and behavioral patterns in two distinct domains : [[ tutorial dialogue ]] -LRB- Kumar et al. , submitted -RRB- and << on-line communities >> -LRB- Arguello et al. , 2006 -RRB- .
CONJUNCTION
[ 30, 31, 40, 41 ]
As evidence of its usefulness and usability , it has been used successfully in a research context to uncover relationships between language and behavioral patterns in two distinct << domains >> : tutorial dialogue -LRB- Kumar et al. , submitted -RRB- and [[ on-line communities ]] -LRB- Arguello et al. , 2006 -RRB- .
HYPONYM-OF
[ 40, 41, 28, 28 ]
As an [[ educational tool ]] , it has been used as part of a unit on << protocol analysis >> in an Educational Research Methods course .
USED-FOR
[ 2, 3, 15, 16 ]
Sources of training data suitable for << language modeling >> of [[ conversational speech ]] are limited .
USED-FOR
[ 9, 10, 6, 7 ]
In this paper , we show how training data can be supplemented with text from the web filtered to match the style and/or topic of the target << recognition task >> , but also that it is possible to get bigger performance gains from the data by using [[ class-dependent interpolation of N-grams ]] .
USED-FOR
[ 46, 49, 27, 28 ]
We present a [[ method ]] for << detecting 3D objects >> using multi-modalities .
USED-FOR
[ 3, 3, 5, 7 ]
We present a << method >> for detecting 3D objects using [[ multi-modalities ]] .
USED-FOR
[ 9, 9, 3, 3 ]
While [[ it ]] is generic , we demonstrate << it >> on the combination of an image and a dense depth map which give complementary object information .
USED-FOR
[ 1, 1, 7, 7 ]
While it is generic , we demonstrate << it >> on the combination of an [[ image ]] and a dense depth map which give complementary object information .
USED-FOR
[ 13, 13, 7, 7 ]
While it is generic , we demonstrate it on the combination of an [[ image ]] and a << dense depth map >> which give complementary object information .
CONJUNCTION
[ 13, 13, 16, 18 ]
While it is generic , we demonstrate << it >> on the combination of an image and a [[ dense depth map ]] which give complementary object information .
USED-FOR
[ 16, 18, 7, 7 ]
While it is generic , we demonstrate it on the combination of an image and a << dense depth map >> which give [[ complementary object information ]] .
FEATURE-OF
[ 21, 23, 16, 18 ]
It is based on an efficient representation of [[ templates ]] that capture the different << modalities >> , and we show in many experiments on commodity hardware that our approach significantly outperforms state-of-the-art methods on single modalities .
USED-FOR
[ 8, 8, 13, 13 ]
It is based on an efficient representation of templates that capture the different modalities , and we show in many experiments on commodity hardware that our [[ approach ]] significantly outperforms << state-of-the-art methods >> on single modalities .
COMPARE
[ 26, 26, 29, 30 ]
It is based on an efficient representation of templates that capture the different modalities , and we show in many experiments on commodity hardware that our [[ approach ]] significantly outperforms state-of-the-art methods on << single modalities >> .
USED-FOR
[ 26, 26, 32, 33 ]
It is based on an efficient representation of templates that capture the different modalities , and we show in many experiments on commodity hardware that our approach significantly outperforms [[ state-of-the-art methods ]] on << single modalities >> .
USED-FOR
[ 29, 30, 32, 33 ]
The [[ compact description of a video sequence ]] through a single image map and a dominant motion has applications in several << domains >> , including video browsing and retrieval , compression , mosaicing , and visual summarization .
USED-FOR
[ 1, 6, 20, 20 ]
The << compact description of a video sequence >> through a single [[ image map ]] and a dominant motion has applications in several domains , including video browsing and retrieval , compression , mosaicing , and visual summarization .
USED-FOR
[ 10, 11, 1, 6 ]
The compact description of a video sequence through a single [[ image map ]] and a << dominant motion >> has applications in several domains , including video browsing and retrieval , compression , mosaicing , and visual summarization .
CONJUNCTION
[ 10, 11, 14, 15 ]

No dataset card yet

Downloads last month
27

Models trained or fine-tuned on zapsdcn/sciie