A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Filters
From Biology To Language Change And Diversity
[chapter]
2017
Zenodo
Preserved Fulltext
; Nick Enfield, the participants in the "Dependencies in Language" Workshop in Ardennes, June 2014, and the organizers and participants to the "Causality in the Language Sciences" Workshop in Leipzig, ...
Rick Janssen's more general contributions to the G[ɜ]bils project; Carly Jaques for invaluable help with ArtiVark; Alexandra Dima for illuminating discussions and pointers to the literature concerning causality ...
Of course, their restricted distribution can simply be a statistical fluctuation expected to obtain when enough features are considered, even in the case where there is a bias against clicks due to properties ...
doi:10.5281/zenodo.573779
fatcat:ar7m4ukjhrhadk7psj56c5erh4
VAENAR-TTS: Variational Auto-Encoder based Non-AutoRegressive Text-to-Speech Synthesis
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Furthermore, hard alignment based on phoneme expansion can degrade the naturalness of the synthesized speech. ...
In contrast, the proposed model of VAENAR-TTS is an end-to-end approach that does not require phoneme-level durations. ...
Acknowledgements: This work is partially supported by National Key R&D Program of China (2020AAA0104500), National Natural Science Foundation of China (NSFC) (62076144), the Major Project of National Social ...
arXiv:2107.03298v1
fatcat:ilfpra3r7vezjfmk4taw7ncvg4
Causal Reasoning Meets Visual Representation Learning: A Prospective Study
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
The majority of the existing methods tend to fit the original data/variable distributions and ignore the essential causal relations behind the multi-modal knowledge, which lacks an unified guidance and ...
Due to the emergence of huge amounts of multi-modal heterogeneous spatial/temporal/spatial-temporal data in big data era, the lack of interpretability, robustness, and out-of-distribution generalization ...
Acknowledgments This work was supported by National Natural Science Foundation of China (No. 62002395), and the National Natural Science Foundation of Guangdong Province (No. 2021A15150123). ...
arXiv:2204.12037v4
fatcat:fgrz2vh42bdozcb5lfegfv5x34
Making "fetch" happen: The influence of social and linguistic context on nonstandard word growth and decline
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
Other Versions
We find that dissemination across many linguistic contexts is a sign of growth: words that appear in more linguistic contexts grow faster and survive longer. ...
But unlike other types of innovation, language change is shaped and constrained by the system in which it takes part. ...
Acknowledgments The authors thank the anonymous reviewers, the audience at NWAV 46 for discussion of an early presentation of this study, and the members of Georgia Tech's Computational Linguistics Lab ...
arXiv:1709.00345v4
fatcat:bk3uno743fhyrfxhdp62s732we
Linguistic tone is related to the population frequency of the adaptive haplogroups of two brain size genes, ASPM and Microcephalin
2007
Proceedings of the National Academy of Sciences of the United States of America
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
The relationship between genetic and linguistic diversity in this case may be causal: certain alleles can bias language acquisition or processing and thereby influence the trajectory of language change ...
Here, we consider the relation between allele frequencies and linguistic typological features. ...
For each linguistic feature, the number of alleles correlating with it in the top 5% of the empirical distribution across the chromosomes does not deviate from the expected distribution ( 2 tests are nonsignificant ...
doi:10.1073/pnas.0610848104
pmid:17537923
pmcid:PMC1904158
fatcat:agjzmh2vengdpkaj6wxo2abpj4
Making "fetch" happen: The influence of social and linguistic context on nonstandard word growth and decline
2018
Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Dissemination across many linguistic contexts is a predictor of success: words that appear in more linguistic contexts grow faster and survive longer. ...
To investigate the role of social and structural factors in language change, we undertake a large-scale analysis of the frequencies of nonstandard words in Reddit. ...
Acknowledgments The authors thank the anonymous reviewers, the audience at NWAV 46 for discussion of an early presentation of this study, and the members of Georgia Tech's Computational Linguistics Lab ...
doi:10.18653/v1/d18-1467
dblp:conf/emnlp/StewartE18
fatcat:wb53pao5pnahxakrahj6ha45um
Universal typological dependencies should be detectable in the history of language families
2011
Linguistic Typology
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Acknowledgements: We would like to thank Frans Plank for giving us this opportunity to engage in this public discussion of issues that we believe are important to the future of linguistic typology. ...
The distribution of states in (b) is the product of three pairs of correlated changes, and the probability that this reveals a causal mechanism linking these states is much higher. ...
the three pairs of changes which produced the distribution of features states shown in (b) . ...
doi:10.1515/lity.2011.034
fatcat:3gse6ffzhncszkxzxb27qjbwbi
Making "fetch" happen: The influence of social and linguistic context on nonstandard word growth and decline
[article]
2018
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2019; you can also visit the original URL.
The file type is application/pdf.
Note that this fulltext copy is not of the "primary" version of this work. The version it corresponds to is:
2018 pre-print arXiv:1802.04140v1 fatcat:xt7xebsw7bdgpp22bo6dklvpoy
We find that dissemination across many linguistic contexts is a sign of growth: words that appear in more linguistic contexts grow faster and survive longer. ...
But unlike other types of innovation, language change is shaped and constrained by the system in which it takes part. ...
Both of these results mirror the finding from the causal inference analysis, that higher linguistic and subreddit dissemination lead to higher probability of growth. ...
arXiv:1802.04140v2
fatcat:mysnwdu4e5hgjacyadckt4tcq4
Language, Economic Development and Economic Globalization: The Case of OECD Countries
2020
Journal of Literature, Languages and Linguistics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2020; you can also visit the original URL.
The file type is application/pdf.
According to the panel causality test results of Emirmahmutoğlu and Köse (2011) , it was found that there is a bidirectional causality relationship between linguistic diversity and economic development ...
in OECD countries, a bidirectional causality relationship between economic development and economic globalization, and a unidirectional causality relationship from linguistic diversity to economic globalization ...
Fisher statistics has an asymptotic chi-square distribution with 2N degrees of independence. In the last stage of the analysis, the direction of causality between variables is investigated. ...
doi:10.7176/jlll/71-07
fatcat:4hdlq2xdjjblllbkbnogaonk7y
Critical Analysis of Deconfounded Pretraining to Improve Visio-Linguistic Models
2022
Frontiers in Artificial Intelligence
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
We provide an explicit explanation of the relation between AutoDeconfounding and the underlying causal model on which it implicitly operates, and show that additional assumptions are needed before the ...
An important problem with many current visio-linguistic models is that they often depend on spurious correlations. ...
In this way, the probability distributions of the Ei , along with Each of these factorizations uses the same number of
the causal mechanisms fi , generate a probability distribution of parameters ...
doi:10.3389/frai.2022.736791
pmid:35402901
pmcid:PMC8993511
fatcat:stbufg6l55h2rhvq7qghfazdlq
Assem-VC: Realistic Voice Conversion by Assembling Modern Speech Synthesis Techniques
[article]
2021
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Recent works on voice conversion (VC) focus on preserving the rhythm and the intonation as well as the linguistic content. ...
We analyze each module with several experiments and reassemble the best components to propose Assem-VC, a new state-of-the-art any-to-many non-parallel VC system. ...
., Yoonhyung Lee from Seoul National University for providing beneficial feedback on the initial version of this paper. ...
arXiv:2104.00931v2
fatcat:yxvyss4zireutmcquc4pkmpniy
The Balkan sprachbund in typological-geographical space
2020
Acta Linguistica Petropolitana
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2021; you can also visit the original URL.
The file type is application/pdf.
They capture some of the classic Balkan features and some deep-seated typological traits, and they enable us to place the Balkan area in the larger typological and linguistic-geographical map of Eurasia ...
The typological properties compared are six complex typological macrofeatures each consisting of a set of related features. ...
Introduction Typological properties are distributed non-randomly in space, and they fi gure in defi nitions of language areas and macroareas, the Balkan sprachbund being probably the fi rst linguistic ...
doi:10.30842/alp2306573716211
fatcat:t6edcwqjyfftbixn72qvkjzpr4
Visual Superordinate Abstraction for Robust Concept Learning
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Although promising progress has been made, existing concept learners are still vulnerable to attribute perturbations and out-of-distribution compositions during inference. ...
linguistic hierarchy. ...
For an object v i in O space, the learner infers the probability of v i sharing a concept corresponding to X by Eq. ( 3 ), which gives the probability distribution p(v i has c x ) for all c x →X. ...
arXiv:2205.14444v1
fatcat:5ykxpk2hsnfydacniexr3h77lq
Informativeness and Invariance: Two Perspectives on Spurious Correlations in Natural Language
[article]
2022
arXiv
pre-print
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2022; you can also visit the original URL.
The file type is application/pdf.
Other Versions
Linking the toy example to a structured causal model shows that (1) feature-label correlations can arise even when the label is invariant to interventions on the feature, and (2) feature-label correlations ...
Gardner et al (2021) argue that the compositional nature of language implies that all correlations between labels and individual "input features" are spurious. ...
The selection of input features is a property of the model and not the dataset, but the intuitive support for UIF seems stronger for features that are lower on the linguistic hierarchy. ...
arXiv:2204.04487v2
fatcat:fuzvcz2jeveevnxga3ipccilyq
Lifecycle of a probabilistic construction
2014
Theoretical Linguistics
Preserved Fulltext
A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf.
Simulation results show that once the model has received enough input, it can predict an intuitive probability distribution over the semantic features. ...
A new construction has a relatively low prior probability (since it only contains one potential frame), and a uniformly distributed conditional probability (since any feature value is equally likely to ...
doi:10.1515/tl-2014-0002
fatcat:fqpitzmmobaqzp4o4mcm37b5ri
« Previous
Showing results 1 — 15 out of 34,281 results