Under statistical learning framework, the paper focuses on how to use traditional linguistic findings on anaphora resolution as a guide for mining and organizing contextual features for Chinese co-reference resolution...Under statistical learning framework, the paper focuses on how to use traditional linguistic findings on anaphora resolution as a guide for mining and organizing contextual features for Chinese co-reference resolution. The main achievements are as follows. (1) In order to simulate "syntactic and semantic parallelism factor", we extract "bags of word form and POS" feature and "bag of seines" feature from the contexts of the entity mentions and incorporate them into the baseline feature set. (2) Because it is too coarse to use the feature of bags of word form, POS tag and seme to determine the syntactic and semantic parallelism between two entity mentions, we propose a method for contextual feature reconstruction based on semantic similarity computation, in order that the reconstructed contextual features could better approximate the anaphora resolution factor of "Syntactic and Semantic Parallelism Preferences". (3) We use an entity-mention-based contextual feature representation instead of isolated word-based contextual feature representation, and expand the size of the contextual windows in addition, in order to approximately simulate "the selectional restriction factor" for anaphora resolution. The experiments show that the multi-level contextual features are useful for co-reference resolution, and the statistical system incorporated with these features performs well on the standard ACE datasets.展开更多
The high-resolution and nondestructive co-reference measurement of the inner and outer threedimensional(3D)surface profiles of laser fusion targets is difficult to achieve.In this study,we propose a laser differential...The high-resolution and nondestructive co-reference measurement of the inner and outer threedimensional(3D)surface profiles of laser fusion targets is difficult to achieve.In this study,we propose a laser differential confocal(LDC)–atomic force probe(AFP)method to measure the inner and outer 3D surface profiles of laser fusion targets at a high resolution.This method utilizes the LDC method to detect the deflection of the AFP and exploits the high spatial resolution of the AFP to enhance the spatial resolution of the outer profile measurement.Nondestructive and co-reference measurements of the inner profile of a target were achieved using the tomographic characteristics of the LDC method.Furthermore,by combining multiple repositionings of the target using a horizontal slewing shaft,the inner and outer 3D surface profiles of the target were obtained,along with a power spectrum assessment of the entire surface.The experimental results revealed that the respective axial and lateral resolutions of the outer profile measurement were 0.5 and 1.3 nm,while the respective axial and lateral resolutions of the inner profile measurement were 2.0 nm and approximately 400.0 nm.The repeatabilities of the rootmean-square deviation measurements for the outer and inner profiles of the target were 2.6 and 2.4 nm,respectively.We believe our study provides a promising method for the high-resolution and nondestructive co-reference measurement of the inner and outer 3D profiles of laser fusion targets.展开更多
Word embedding has drawn a lot of attention due to its usefulness in many NLP tasks. So far a handful of neural-network based word embedding algorithms have been proposed without considering the effects of pronouns in...Word embedding has drawn a lot of attention due to its usefulness in many NLP tasks. So far a handful of neural-network based word embedding algorithms have been proposed without considering the effects of pronouns in the training corpus. In this paper, we propose using co-reference resolution to improve the word embedding by extracting better context. We evaluate four word embeddings with considerations of co-reference resolution and compare the quality of word embedding on the task of word analogy and word similarity on multiple data sets.Experiments show that by using co-reference resolution, the word embedding performance in the word analogy task can be improved by around 1.88%. We find that the words that are names of countries are affected the most,which is as expected.展开更多
Visual Dialog is a multi-modal task involving both computer vision and dialog systems.The goal is to answer multiple questions in conversation style,given an image as the context.Neural networks with attention modules...Visual Dialog is a multi-modal task involving both computer vision and dialog systems.The goal is to answer multiple questions in conversation style,given an image as the context.Neural networks with attention modules are widely used for this task,because of their effectiveness in reasoning the relevance between the texts and images.In this work,we study how to further improve the quality of such reasoning,which is an open challenge.Our baseline is the Recursive Visual Attention(RVA)model,which refines the vision-text attention by iteratively visiting the dialog history.Building on top of that,we propose to improve the attention mechanism with contrastive learning.We train a Matching-Aware Attention Kernel(MAAK)by aligning the deep feature embeddings of an image and its caption,to provide better attention scores.Experiments show consistent improvements from MAAK.In addition,we study the effect of using Multimodal Compact Bilinear(MCB)pooling as a three-way feature fusion for the visual,textual and dialog history embeddings.We analyze the performance of both methods in the discussion section,and propose further ideas to resolve current limitations.展开更多
基金Supported by the National Natural Science Foundation of China under Grant Nos.60372016,60121302,60673042the National High Technology Development 863 Program of China under Grant No.2006AA01Z144the Natural Science Foundation of Beijing under Grant No.4052027.
文摘Under statistical learning framework, the paper focuses on how to use traditional linguistic findings on anaphora resolution as a guide for mining and organizing contextual features for Chinese co-reference resolution. The main achievements are as follows. (1) In order to simulate "syntactic and semantic parallelism factor", we extract "bags of word form and POS" feature and "bag of seines" feature from the contexts of the entity mentions and incorporate them into the baseline feature set. (2) Because it is too coarse to use the feature of bags of word form, POS tag and seme to determine the syntactic and semantic parallelism between two entity mentions, we propose a method for contextual feature reconstruction based on semantic similarity computation, in order that the reconstructed contextual features could better approximate the anaphora resolution factor of "Syntactic and Semantic Parallelism Preferences". (3) We use an entity-mention-based contextual feature representation instead of isolated word-based contextual feature representation, and expand the size of the contextual windows in addition, in order to approximately simulate "the selectional restriction factor" for anaphora resolution. The experiments show that the multi-level contextual features are useful for co-reference resolution, and the statistical system incorporated with these features performs well on the standard ACE datasets.
基金supported by the National Natural Science Foundation of China(52327806 and U22A6006).
文摘The high-resolution and nondestructive co-reference measurement of the inner and outer threedimensional(3D)surface profiles of laser fusion targets is difficult to achieve.In this study,we propose a laser differential confocal(LDC)–atomic force probe(AFP)method to measure the inner and outer 3D surface profiles of laser fusion targets at a high resolution.This method utilizes the LDC method to detect the deflection of the AFP and exploits the high spatial resolution of the AFP to enhance the spatial resolution of the outer profile measurement.Nondestructive and co-reference measurements of the inner profile of a target were achieved using the tomographic characteristics of the LDC method.Furthermore,by combining multiple repositionings of the target using a horizontal slewing shaft,the inner and outer 3D surface profiles of the target were obtained,along with a power spectrum assessment of the entire surface.The experimental results revealed that the respective axial and lateral resolutions of the outer profile measurement were 0.5 and 1.3 nm,while the respective axial and lateral resolutions of the inner profile measurement were 2.0 nm and approximately 400.0 nm.The repeatabilities of the rootmean-square deviation measurements for the outer and inner profiles of the target were 2.6 and 2.4 nm,respectively.We believe our study provides a promising method for the high-resolution and nondestructive co-reference measurement of the inner and outer 3D profiles of laser fusion targets.
基金supported by the National HighTech Research and Development(863)Program(No.2015AA015401)the National Natural Science Foundation of China(Nos.61533018 and 61402220)+2 种基金the State Scholarship Fund of CSC(No.201608430240)the Philosophy and Social Science Foundation of Hunan Province(No.16YBA323)the Scientific Research Fund of Hunan Provincial Education Department(Nos.16C1378 and 14B153)
文摘Word embedding has drawn a lot of attention due to its usefulness in many NLP tasks. So far a handful of neural-network based word embedding algorithms have been proposed without considering the effects of pronouns in the training corpus. In this paper, we propose using co-reference resolution to improve the word embedding by extracting better context. We evaluate four word embeddings with considerations of co-reference resolution and compare the quality of word embedding on the task of word analogy and word similarity on multiple data sets.Experiments show that by using co-reference resolution, the word embedding performance in the word analogy task can be improved by around 1.88%. We find that the words that are names of countries are affected the most,which is as expected.
文摘Visual Dialog is a multi-modal task involving both computer vision and dialog systems.The goal is to answer multiple questions in conversation style,given an image as the context.Neural networks with attention modules are widely used for this task,because of their effectiveness in reasoning the relevance between the texts and images.In this work,we study how to further improve the quality of such reasoning,which is an open challenge.Our baseline is the Recursive Visual Attention(RVA)model,which refines the vision-text attention by iteratively visiting the dialog history.Building on top of that,we propose to improve the attention mechanism with contrastive learning.We train a Matching-Aware Attention Kernel(MAAK)by aligning the deep feature embeddings of an image and its caption,to provide better attention scores.Experiments show consistent improvements from MAAK.In addition,we study the effect of using Multimodal Compact Bilinear(MCB)pooling as a three-way feature fusion for the visual,textual and dialog history embeddings.We analyze the performance of both methods in the discussion section,and propose further ideas to resolve current limitations.