In his [2011], Lionel Shapiro proposes a very interesting deflationist conception of logical consequence, according to which the logical-consequence predicate applying to sentences serves as a merely expressive device allowing us to talk in a generalising way about typically non-linguistic entailment facts (just as, according to a deflationist conception of truth, the truth predicate applying t...

In recent years, neural machine translation (NMT) has demonstrated state-of-the-art machine translation (MT) performance. It is a new approach to MT, which tries to learn a set of parameters to maximize the conditional probability of target sentences given source sentences. In this paper, we present a novel approach to improve the translation performance in NMT by conveying topic knowledge duri...

background: cerebrovascular disease leading to stroke is the most common cause of aphasia. speakers with agrammatic non-fluent aphasia have difficulties in production of movement-derived sentences such as passive sentences, topicalized constituents, and wh-questions. to assess the production of complex sentences, some passive, topicalized and focused sentences were designed for patients with no...

Clemens Boucsein,* Martin Nawrot,* Stefan Rotter, Ad Aertsen, and Detlef Heck Neurobiology and Biophysics, Institute of Biology III and Bernstein Center for Computational Neuroscience Freiburg, Albert-Ludwigs-University, Freiburg; Institute for Frontier Areas of Psychology and Mental Health, Freiburg, Germany; and University of Tennessee Health Science Center, Department of Anatomy and Neurobio...

We present a symbolic machinery that admits both probabilistic and causal information about a given domain and produces probabilistic statements about he effect of actions and the impact of observations. Thecalculus admits two types of conditioning operators: ordinary Bayes conditioning, P(y]X = z), which represents he observation X z, and causal conditioning, P(yldo(X = x)), read the probabili...

Category text generation receives considerable attentions since it is beneficial for various natural language processing tasks. Recently, the generative adversarial network (GAN) has attained promising performance in generation, attributed to its training process. However, there are several issues GANs, including discreteness, instability, mode collapse, lack of diversity and controllability et...

The question “Does parole work?” can be given a number of meanings. One is whether parole—by which I mean the early release by a parole board of a prisoner on license so that a shorter period of the sentence is served in prison and a longer proportion under supervision—has a beneficial impact on the probability of recidivism: Is there, as it is sometimes put, a “parole effect”? Another question...

This paper presents an effective method for generating natural language sentences from their underlying meaning representations. The method is built on top of a hybrid tree representation that jointly encodes both the meaning representation as well as the natural language in a tree structure. By using a tree conditional random field on top of the hybrid tree representation, we are able to expli...