Filters








32,773 Hits in 4.7 sec

Joint Semantic Analysis with Document-Level Cross-Task Coherence Rewards [article]

Rahul Aralikatte, Mostafa Abdou, Heather Lent, Daniel Hershcovich, Anders Søgaard
2020 arXiv   pre-print
We present a neural network architecture for joint coreference resolution and semantic role labeling for English, and train graph neural networks to model the 'coherence' of the combined shallow semantic  ...  Coreference resolution and semantic role labeling are NLP tasks that capture different aspects of semantics, indicating respectively, which expressions refer to the same entity, and what semantic roles  ...  We explicitly model their coherence, presenting a reinforcement learning architecture for semi-supervised finetuning of coreference resolvers and semantic role labelers with coherence rewards on unlabeled  ... 
arXiv:2010.05567v1 fatcat:3nc6abqx3zgnlbyqxtbj2gtmxa

Labeling Discourse to Build Academic Persona

Geneviève Bordet
2015 Journal of Academic Writing  
Labeling nouns are identified and semantically classified for each discipline, according to linguistic origin. Case studies show that success requires adequate lexical choice of labeling nouns.  ...  This study focuses on the use of general 'labeling' nouns as a factor of coherence and rhetorical persuasion in scientific abstracts, with specific interest in terms that are determined by an anaphoric  ...  Nouns as a cohesive and interpretive device A CONTRASTIVE APPROACH My approach to the cohesive and interpretive role of the labeling noun determined by 'this' is based on a study of PhD abstracts written  ... 
doi:10.18552/joaw.v5i1.165 fatcat:kevustb3dvfdhict7kqfpryfl4

Automatically Identifying Implicit Arguments to Improve Argument Linking and Coherence Modeling

Michael Roth, Anette Frank
2013 Joint Conference on Lexical and Computational Semantics  
as training data for improving existing argument linking models; finally, we present a novel approach to modeling local coherence that extends previous approaches by taking into account non-explicit entity  ...  Implicit arguments are a discourse-level phenomenon that has not been extensively studied in semantic processing. One reason for this lies in the scarce amount of annotated data sets available.  ...  Acknowledgments We are grateful to the Landesgraduiertenförderung Baden-Württemberg for funding within the research initiative "Coherence in language processing" at Heidelberg University.  ... 
dblp:conf/starsem/RothF13 fatcat:q6skhaenlvhinpjdqtzajeqawi

An Automatic Linguistics Approach for Persian Document Summarization

Hossein Kamyar, Mohsen Kahani, Mohsen Kamyar, Asef Poormasoomi
2011 2011 International Conference on Asian Language Processing  
This pipeline consists of several components such as co-reference resolution, semantic role labeling and POS [Part of speech] tagging.  ...  __ In this paper we propose a novel technique for summarizing a text based on the linguistics properties of text elements and semantic chains among them.  ...  Acknowledgment This work was supported by the Web Technology Laboratory of Ferdowsi University of Mashhad. We would like to thank WTLAB group members.  ... 
doi:10.1109/ialp.2011.52 dblp:conf/ialp/KamyarKKP11 fatcat:u6bgg7umozf4tib4jyxe4lkuuy

Joint Semantic Analysis with Document-Level Cross-Task Coherence Rewards

Rahul Aralikatte, Mostafa Abdou, Heather C Lent, Daniel Hershcovich, Anders Søgaard
2021 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
We present a neural network architecture for joint coreference resolution and semantic role labeling for English, and train graph neural networks to model the 'coherence' of the combined shallow semantic  ...  Coreference resolution and semantic role labeling are NLP tasks that capture different aspects of semantics, indicating respectively, which expressions refer to the same entity, and what semantic roles  ...  We explicitly model their coherence, presenting a reinforcement learning architecture for semi-supervised finetuning of coreference resolvers and semantic role labelers with coherence rewards on unlabeled  ... 
doi:10.1609/aaai.v35i14.17484 fatcat:lw5mrhzmcjd4jd4nb62wcl7zdy

Inducing Implicit Arguments from Comparable Texts: A Framework and Its Applications

Michael Roth, Anette Frank
2015 Computational Linguistics  
Our framework builds upon a novel projection approach that allows for the accurate detection of implicit arguments by aligning and comparing predicate-argument structures across pairs of comparable texts  ...  As part of this framework, we develop a graph-based model for predicate alignment that significantly outperforms previous approaches.  ...  We are grateful to the anonymous reviewers for helpful feedback and suggestions.  ... 
doi:10.1162/coli_a_00236 fatcat:2kqvlb7nifdmjl75x6wv7tb4zq

Research on Discourse Coherence based on the Analysis Model of Event Chain from the Perspective of Computational Linguistics

Min Fan, Shanwen Xu, N. Hamid, S. Zhou
2020 E3S Web of Conferences  
Based on the event chain of the text semantic structure representation, this paper proposes a text semantic structure representation model, on the basis of which, text coherent resources can be used for  ...  the task of text semantic analysis.  ...  assumptions of entity-based discourse coherence models, while using the existing mature sentences and event analysis technology to provide a new approach for the determinating and applying of discourse  ... 
doi:10.1051/e3sconf/202018903025 fatcat:fv4brf3cubh5bnwzslkexs7a7m

Unsupervised Entity Linking with Abstract Meaning Representation

Xiaoman Pan, Taylor Cassidy, Ulf Hermjakob, Heng Ji, Kevin Knight
2015 Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies  
hand labeled traditional semantic roles as context representation for EL.  ...  Finally, we show promising preliminary results for using AMR to select sets of "coherent" entity mentions for collective entity linking 1 .  ...  Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation here on.  ... 
doi:10.3115/v1/n15-1119 dblp:conf/naacl/PanCHJK15 fatcat:go37vvtocvfa5p2eho3jw2z3yy

Using Domain Information for Word Sense Disambiguation

Bernardo Magnini, Carlo Strapparava, Giovanni Pezzulo, Alfio Gliozzo
2001 International Workshop on Semantic Evaluation  
The major goal in ITC-irst's participation at SENSEVAL-2 was to test the role of domain information in word sense disambiguation.  ...  The underlying working hypothesis is that domain labels, such as MEDICINE, ARCHITECTURE and SPORT provide a natural way to establish semantic relations among word senses, which can be profitably used during  ...  For SENSEVAL-2 the goal was to evaluate the role of domain information in WSD: no other syntactic or semantic information has been used (e.g. semantic relations in WoRDNET) except domain labels.  ... 
dblp:conf/semeval/MagniniSPG01 fatcat:ebemgribober7p6nqxqn44j7g4

Unsupervised Learning of Coherent and General Semantic Classes for Entity Aggregates

Henry Anaya-Sánchez, Anselmo Peñas
2015 International Conference on Computational Semantics  
as a particular semantic role in a collection of text documents).  ...  Thus, the set of learned classes is then used to devise a generative model for entity categorization tasks such as semantic class induction.  ...  Acknowledgments This work was partially funded by MINECO (PCIN-2013-002-C02-01) and EPSRC (EP/K017845/1) in the framework of CHIST-ERA READERS project, and by project Voxpopuli (TIN2013-47090-C3-1).  ... 
dblp:conf/iwcs/Anaya-SanchezP15 fatcat:nqtf2igvmvfffcbynflkgi3dui

Using semantic roles to improve summaries

Diana Trandabat
2011 European Workshop on Natural Language Generation  
; thirdly, selection of the sentences containing specific semantic roles for the most relevant entities in text.  ...  This paper describes preliminary analysis on the influence of the semantic roles in summary generation.  ...  Acknowledgments The research presented in this paper was funded by the Sectoral Operational Programme for Human Resources Development through the project "Development of the innovation capacity and increasing  ... 
dblp:conf/enlg/Trandabat11 fatcat:dj2mex5zizclhkehs4bszn66ja

Semantic Role Labeling: Past, Present and Future

Lluís Màrquez
2009 Annual Meeting of the Association for Computational Linguistics  
Introduction Semantic Role Labeling (SRL) consists of, given a sentence, detecting basic event structures such as "who" did "what" to "whom", "when" and "where".  ...  Keys for future progress • Relation to syntax: joint learning of syntactic and semantic dependencies • Generalization across domains and text genres • Use of semantic knowledge • SRL systems in applications  ... 
dblp:conf/acl/Marquez09 fatcat:5kqn52ayhrh63cd3uyql426cgy

Semantic Frame Parsing for Information Extraction : the CALOR corpus [article]

Gabriel Marzinotto, Frederic Bechet
2018 arXiv   pre-print
The approaches compared in this study rely on an integrated sequence labeling model which jointly optimizes frame identification and semantic role segmentation and identification.  ...  Such an approach leverages the manual annotation of larger corpora than those obtained through full text parsing and therefore opens the door to alternative methods for Frame parsing than those used so  ...  Even though our approaches perform frame detection and semantic role labeling in one step, we detail the scores for each sub-task because they are relevant indicators of the performance of a parser, and  ... 
arXiv:1812.08039v1 fatcat:hvo3iqujprfd5ngj24madkjksy

Dialogue Coherence Assessment Without Explicit Dialogue Act Labels [article]

Mohsen Mesgar, Sebastian Bücker, Iryna Gurevych
2020 arXiv   pre-print
Our approach alleviates the need for explicit dialogue act labels during evaluation.  ...  It indicates two drawbacks, (a) semantics of utterances is limited to entity mentions, and (b) the performance of coherence models strongly relies on the quality of the input dialogue act labels.  ...  We also thank anonymous reviewers for their useful suggestions for improving the quality of the paper.  ... 
arXiv:1908.08486v2 fatcat:rocizcurfjbr7fwxz2rpegffoi

Automatic Evaluation of Text Coherence: Models and Representations

Mirella Lapata, Regina Barzilay
2005 International Joint Conference on Artificial Intelligence  
This paper investigates the automatic evaluation of text coherence for machine-generated texts.  ...  Our modeling approach relies on shallow text properties and is relatively inexpensive.  ...  Acknowledgments The authors acknowledge the support of EPSRC (Lapata; grant GR/T04540/01) and the National Science Foundation (Barzilay; CAREER grant IIS-0448168).  ... 
dblp:conf/ijcai/LapataB05 fatcat:7bqdeaizp5gkpm7d45wrm53uju
« Previous Showing results 1 — 15 out of 32,773 results