Survey of the state of the art in natural language generation: Core tasks, applications and evaluation
This paper surveys the current state of the art in Natural Language Generation (NLG),
defined as the task of generating text or speech from non-linguistic input. A survey of NLG is …
defined as the task of generating text or speech from non-linguistic input. A survey of NLG is …
Measuring compositionality in representation learning
Many machine learning algorithms represent input data with vector embeddings or discrete
codes. When inputs exhibit compositional structure (eg objects built from parts or procedures …
codes. When inputs exhibit compositional structure (eg objects built from parts or procedures …
Computational generation of referring expressions: A survey
This article offers a survey of computational research on referring expression generation
(REG). It introduces the REG problem and describes early work in this area, discussing what …
(REG). It introduces the REG problem and describes early work in this area, discussing what …
Reasoning about pragmatics with neural listeners and speakers
We present a model for pragmatically describing scenes, in which contrastive behavior
results from a combination of inference-driven pragmatics and learned semantics. Like …
results from a combination of inference-driven pragmatics and learned semantics. Like …
[ספר][B] Computational models of referring: a study in cognitive science
K Van Deemter - 2016 - books.google.com
An argument that computational models can shed light on referring, a fundamental and
much-studied aspect of communication. To communicate, speakers need to make it clear …
much-studied aspect of communication. To communicate, speakers need to make it clear …
Factors causing overspecification in definite descriptions
Speakers often overspecify their target descriptions and include more information than
necessary for unique identification of the target referent. In the current paper, we study the …
necessary for unique identification of the target referent. In the current paper, we study the …
[PDF][PDF] The TUNA challenge 2008: Overview and evaluation results
The TUNA Challenge was a set of three shared tasks at REG'08, all of which used data from
the TUNA Corpus. The three tasks covered attribute selection for referring expressions …
the TUNA Corpus. The three tasks covered attribute selection for referring expressions …
[PDF][PDF] The use of spatial relations in referring expression generation
There is a prevailing assumption in the literature on referring expression generation that
relations are used in descriptions only 'as a last resort', typically on the basis that including …
relations are used in descriptions only 'as a last resort', typically on the basis that including …
The effect of scene variation on the redundant use of color in definite reference
This study investigates to what extent the amount of variation in a visual scene causes
speakers to mention the attribute color in their definite target descriptions, focusing on …
speakers to mention the attribute color in their definite target descriptions, focusing on …
Learning in the rational speech acts model
The Rational Speech Acts (RSA) model treats language use as a recursive process in which
probabilistic speaker and listener agents reason about each other's intentions to enrich the …
probabilistic speaker and listener agents reason about each other's intentions to enrich the …