May 4, 2019

1300 words 7 mins read

Paper Group NANR 165

Joint Word Segmentation and Phonetic Category Induction. Learning Succinct Models: Pipelined Compression with L1-Regularization, Hashing, Elias-Fano Indices, and Quantization. BulPhonC: Bulgarian Speech Corpus for the Development of ASR Technology. SemEval-2016 Task 14: Semantic Taxonomy Enrichment. A Corpus of Word-Aligned Asked and Anticipated Qu …

Joint Word Segmentation and Phonetic Category Induction


Title	Joint Word Segmentation and Phonetic Category Induction
Authors	Micha Elsner, Stephanie Antetomaso, Naomi Feldman
Abstract
Tasks	Language Acquisition, Speech Recognition
Published	2016-08-01
URL	https://www.aclweb.org/anthology/P16-2010/
PDF	https://www.aclweb.org/anthology/P16-2010
PWC	https://paperswithcode.com/paper/joint-word-segmentation-and-phonetic-category
Repo
Framework

Learning Succinct Models: Pipelined Compression with L1-Regularization, Hashing, Elias-Fano Indices, and Quantization


Title	Learning Succinct Models: Pipelined Compression with L1-Regularization, Hashing, Elias-Fano Indices, and Quantization
Authors	Hajime Senuma, Akiko Aizawa
Abstract	The recent proliferation of smart devices necessitates methods to learn small-sized models. This paper demonstrates that if there are $m$ features in total but only $n = o(\sqrt{m})$ features are required to distinguish examples, with $\Omega(\log m)$ training examples and reasonable settings, it is possible to obtain a good model in a \textit{succinct} representation using $n \log_2 \frac{m}{n} + o(m)$ bits, by using a pipeline of existing compression methods: L1-regularized logistic regression, feature hashing, Elias{–}Fano indices, and randomized quantization. An experiment shows that a noun phrase chunking task for which an existing library requires 27 megabytes can be compressed to less than 13 \textit{kilo}bytes without notable loss of accuracy.
Tasks	Chunking, Quantization
Published	2016-12-01
URL	https://www.aclweb.org/anthology/C16-1261/
PDF	https://www.aclweb.org/anthology/C16-1261
PWC	https://paperswithcode.com/paper/learning-succinct-models-pipelined
Repo
Framework

BulPhonC: Bulgarian Speech Corpus for the Development of ASR Technology


Title	BulPhonC: Bulgarian Speech Corpus for the Development of ASR Technology
Authors	Neli Hateva, Petar Mitankin, Stoyan Mihov
Abstract	In this paper we introduce a Bulgarian speech database, which was created for the purpose of ASR technology development. The paper describes the design and the content of the speech database. We present also an empirical evaluation of the performance of a LVCSR system for Bulgarian trained on the BulPhonC data. The resource is available free for scientific usage.
Tasks	Large Vocabulary Continuous Speech Recognition
Published	2016-05-01
URL	https://www.aclweb.org/anthology/L16-1123/
PDF	https://www.aclweb.org/anthology/L16-1123
PWC	https://paperswithcode.com/paper/bulphonc-bulgarian-speech-corpus-for-the
Repo
Framework

SemEval-2016 Task 14: Semantic Taxonomy Enrichment


Title	SemEval-2016 Task 14: Semantic Taxonomy Enrichment
Authors	David Jurgens, Mohammad Taher Pilehvar
Abstract
Tasks	Information Retrieval, Semantic Textual Similarity, Sentiment Analysis, Word Sense Disambiguation
Published	2016-06-01
URL	https://www.aclweb.org/anthology/S16-1169/
PDF	https://www.aclweb.org/anthology/S16-1169
PWC	https://paperswithcode.com/paper/semeval-2016-task-14-semantic-taxonomy
Repo
Framework

A Corpus of Word-Aligned Asked and Anticipated Questions in a Virtual Patient Dialogue System


Title	A Corpus of Word-Aligned Asked and Anticipated Questions in a Virtual Patient Dialogue System
Authors	Ajda Gokcen, Evan Jaffe, Johnsey Erdmann, Michael White, Douglas Danforth
Abstract	We present a corpus of virtual patient dialogues to which we have added manually annotated gold standard word alignments. Since each question asked by a medical student in the dialogues is mapped to a canonical, anticipated version of the question, the corpus implicitly defines a large set of paraphrase (and non-paraphrase) pairs. We also present a novel process for selecting the most useful data to annotate with word alignments and for ensuring consistent paraphrase status decisions. In support of this process, we have enhanced the earlier Edinburgh alignment tool (Cohn et al., 2008) and revised and extended the Edinburgh guidelines, in particular adding guidance intended to ensure that the word alignments are consistent with the overall paraphrase status decision. The finished corpus and the enhanced alignment tool are made freely available.
Tasks
Published	2016-05-01
URL	https://www.aclweb.org/anthology/L16-1506/
PDF	https://www.aclweb.org/anthology/L16-1506
PWC	https://paperswithcode.com/paper/a-corpus-of-word-aligned-asked-and
Repo
Framework

Analyzing Framing through the Casts of Characters in the News


Title	Analyzing Framing through the Casts of Characters in the News
Authors	Dallas Card, Justin Gross, Amber Boydstun, Noah A. Smith
Abstract
Tasks	Model Selection, Topic Models
Published	2016-11-01
URL	https://www.aclweb.org/anthology/D16-1148/
PDF	https://www.aclweb.org/anthology/D16-1148
PWC	https://paperswithcode.com/paper/analyzing-framing-through-the-casts-of
Repo
Framework

Interlocking Phrases in Phrase-based Statistical Machine Translation


Title	Interlocking Phrases in Phrase-based Statistical Machine Translation
Authors	Ye Kyaw Thu, Andrew Finch, Eiichiro Sumita
Abstract
Tasks	Language Modelling, Machine Translation
Published	2016-06-01
URL	https://www.aclweb.org/anthology/N16-1124/
PDF	https://www.aclweb.org/anthology/N16-1124
PWC	https://paperswithcode.com/paper/interlocking-phrases-in-phrase-based
Repo
Framework

CogALex-V Shared Task: Mach5 – A traditional DSM approach to semantic relatedness


Title	CogALex-V Shared Task: Mach5 – A traditional DSM approach to semantic relatedness
Authors	Stefan Evert
Abstract	This contribution provides a strong baseline result for the CogALex-V shared task using a traditional {``}count{''}-type DSM (placed in rank 2 out of 7 in subtask 1 and rank 3 out of 6 in subtask 2). Parameter tuning experiments reveal some surprising effects and suggest that the use of random word pairs as negative examples may be problematic, guiding the parameter optimization in an undesirable direction. \|
Tasks
Published	2016-12-01
URL	https://www.aclweb.org/anthology/W16-5312/
PDF	https://www.aclweb.org/anthology/W16-5312
PWC	https://paperswithcode.com/paper/cogalex-v-shared-task-mach5-a-a-traditional
Repo
Framework

How Naked is the Naked Truth? A Multilingual Lexicon of Nominal Compound Compositionality


Title	How Naked is the Naked Truth? A Multilingual Lexicon of Nominal Compound Compositionality
Authors	Carlos Ramisch, Silvio Cordeiro, Leonardo Zilio, Marco Idiart, Aline Villavicencio
Abstract
Tasks
Published	2016-08-01
URL	https://www.aclweb.org/anthology/P16-2026/
PDF	https://www.aclweb.org/anthology/P16-2026
PWC	https://paperswithcode.com/paper/how-naked-is-the-naked-truth-a-multilingual
Repo
Framework

Detecting Mild Cognitive Impairment by Exploiting Linguistic Information from Transcripts


Title	Detecting Mild Cognitive Impairment by Exploiting Linguistic Information from Transcripts
Authors	Veronika Vincze, G{'a}bor Gosztolya, L{'a}szl{'o} T{'o}th, Ildik{'o} Hoffmann, Gr{'e}ta Szatl{'o}czki, Zolt{'a}n B{'a}nr{'e}ti, Magdolna P{'a}k{'a}ski, J{'a}nos K{'a}lm{'a}n
Abstract
Tasks	Lexical Analysis, Speech Recognition
Published	2016-08-01
URL	https://www.aclweb.org/anthology/P16-2030/
PDF	https://www.aclweb.org/anthology/P16-2030
PWC	https://paperswithcode.com/paper/detecting-mild-cognitive-impairment-by
Repo
Framework

ETE 3: Reconstruction, Analysis, and Visualization of Phylogenomic Data


Title	ETE 3: Reconstruction, Analysis, and Visualization of Phylogenomic Data
Authors	Jaime Huerta-Cepas, François Serra, Peer Bork
Abstract	The Environment for Tree Exploration (ETE) is a computational framework that simplifies the reconstruction, analysis, and visualization of phylogenetic trees and multiple sequence alignments. Here, we present ETE v3, featuring numerous improvements in the underlying library of methods, and providing a novel set of standalone tools to perform common tasks in comparative genomics and phylogenetics. The new features include (i) building gene-based and supermatrix-based phylogenies using a single command, (ii) testing and visualizing evolutionary models, (iii) calculating distances between trees of different size or including duplications, and (iv) providing seamless integration with the NCBI taxonomy database. ETE is freely available at http://etetoolkit.org
Tasks
Published	2016-02-26
URL	https://academic.oup.com/mbe/article/33/6/1635/2579822
PDF	https://academic.oup.com/mbe/article-pdf/33/6/1635/7953632/msw046.pdf
PWC	https://paperswithcode.com/paper/ete-3-reconstruction-analysis-and
Repo
Framework

Do Enterprises Have Emotions?


Title	Do Enterprises Have Emotions?
Authors	Sven Buechel, Udo Hahn, Jan Goldenstein, Sebastian G. M. H{"a}ndschke, Peter Walgenbach
Abstract
Tasks
Published	2016-06-01
URL	https://www.aclweb.org/anthology/W16-0423/
PDF	https://www.aclweb.org/anthology/W16-0423
PWC	https://paperswithcode.com/paper/do-enterprises-have-emotions
Repo
Framework

Deep Learning for Predicting Human Strategic Behavior


Title	Deep Learning for Predicting Human Strategic Behavior
Authors	Jason S. Hartford, James R. Wright, Kevin Leyton-Brown
Abstract	Predicting the behavior of human participants in strategic settings is an important problem in many domains. Most existing work either assumes that participants are perfectly rational, or attempts to directly model each participant’s cognitive processes based on insights from cognitive psychology and experimental economics. In this work, we present an alternative, a deep learning approach that automatically performs cognitive modeling without relying on such expert knowledge. We introduce a novel architecture that allows a single network to generalize across different input and output dimensions by using matrix units rather than scalar units, and show that its performance significantly outperforms that of the previous state of the art, which relies on expert-constructed features.
Tasks
Published	2016-12-01
URL	http://papers.nips.cc/paper/6509-deep-learning-for-predicting-human-strategic-behavior
PDF	http://papers.nips.cc/paper/6509-deep-learning-for-predicting-human-strategic-behavior.pdf
PWC	https://paperswithcode.com/paper/deep-learning-for-predicting-human-strategic
Repo
Framework

ENIAM: Categorial Syntactic-Semantic Parser for Polish


Title	ENIAM: Categorial Syntactic-Semantic Parser for Polish
Authors	Wojciech Jaworski, Jakub Kozakoszczak
Abstract	This paper presents ENIAM, the first syntactic and semantic parser that generates semantic representations for sentences in Polish. The parser processes non-annotated data and performs tokenization, lemmatization, dependency recognition, word sense annotation, thematic role annotation, partial disambiguation and computes the semantic representation.
Tasks	Information Retrieval, Lemmatization, Natural Language Inference, Question Answering, Tokenization
Published	2016-12-01
URL	https://www.aclweb.org/anthology/C16-2051/
PDF	https://www.aclweb.org/anthology/C16-2051
PWC	https://paperswithcode.com/paper/eniam-categorial-syntactic-semantic-parser
Repo
Framework

LAMB: A Good Shepherd of Morphologically Rich Languages


Title	LAMB: A Good Shepherd of Morphologically Rich Languages
Authors	Sebastian Ebert, Thomas M{"u}ller, Hinrich Sch{"u}tze
Abstract
Tasks	Lemmatization
Published	2016-11-01
URL	https://www.aclweb.org/anthology/D16-1071/
PDF	https://www.aclweb.org/anthology/D16-1071
PWC	https://paperswithcode.com/paper/lamb-a-good-shepherd-of-morphologically-rich
Repo
Framework