@inproceedings{suglia-etal-2020-imagining,
title = "Imagining Grounded Conceptual Representations from Perceptual Information in Situated Guessing Games",
author = "Suglia, Alessandro and
Vergari, Antonio and
Konstas, Ioannis and
Bisk, Yonatan and
Bastianelli, Emanuele and
Vanzo, Andrea and
Lemon, Oliver",
editor = "Scott, Donia and
Bel, Nuria and
Zong, Chengqing",
booktitle = "Proceedings of the 28th International Conference on Computational Linguistics",
month = dec,
year = "2020",
address = "Barcelona, Spain (Online)",
publisher = "International Committee on Computational Linguistics",
url = "https://aclanthology.org/2020.coling-main.95",
doi = "10.18653/v1/2020.coling-main.95",
pages = "1090--1102",
abstract = "In visual guessing games, a Guesser has to identify a target object in a scene by asking questions to an Oracle. An effective strategy for the players is to learn conceptual representations of objects that are both discriminative and expressive enough to ask questions and guess correctly. However, as shown by Suglia et al. (2020), existing models fail to learn truly multi-modal representations, relying instead on gold category labels for objects in the scene both at training and inference time. This provides an unnatural performance advantage when categories at inference time match those at training time, and it causes models to fail in more realistic {``}zero-shot{''} scenarios where out-of-domain object categories are involved. To overcome this issue, we introduce a novel {``}imagination{''} module based on Regularized Auto-Encoders, that learns context-aware and category-aware latent embeddings without relying on category labels at inference time. Our imagination module outperforms state-of-the-art competitors by 8.26{\%} gameplay accuracy in the CompGuessWhat?! zero-shot scenario (Suglia et al., 2020), and it improves the Oracle and Guesser accuracy by 2.08{\%} and 12.86{\%} in the GuessWhat?! benchmark, when no gold categories are available at inference time. The imagination module also boosts reasoning about object properties and attributes.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="suglia-etal-2020-imagining">
<titleInfo>
<title>Imagining Grounded Conceptual Representations from Perceptual Information in Situated Guessing Games</title>
</titleInfo>
<name type="personal">
<namePart type="given">Alessandro</namePart>
<namePart type="family">Suglia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Vergari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ioannis</namePart>
<namePart type="family">Konstas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yonatan</namePart>
<namePart type="family">Bisk</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Emanuele</namePart>
<namePart type="family">Bastianelli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrea</namePart>
<namePart type="family">Vanzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Oliver</namePart>
<namePart type="family">Lemon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 28th International Conference on Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Donia</namePart>
<namePart type="family">Scott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nuria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chengqing</namePart>
<namePart type="family">Zong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>International Committee on Computational Linguistics</publisher>
<place>
<placeTerm type="text">Barcelona, Spain (Online)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In visual guessing games, a Guesser has to identify a target object in a scene by asking questions to an Oracle. An effective strategy for the players is to learn conceptual representations of objects that are both discriminative and expressive enough to ask questions and guess correctly. However, as shown by Suglia et al. (2020), existing models fail to learn truly multi-modal representations, relying instead on gold category labels for objects in the scene both at training and inference time. This provides an unnatural performance advantage when categories at inference time match those at training time, and it causes models to fail in more realistic “zero-shot” scenarios where out-of-domain object categories are involved. To overcome this issue, we introduce a novel “imagination” module based on Regularized Auto-Encoders, that learns context-aware and category-aware latent embeddings without relying on category labels at inference time. Our imagination module outperforms state-of-the-art competitors by 8.26% gameplay accuracy in the CompGuessWhat?! zero-shot scenario (Suglia et al., 2020), and it improves the Oracle and Guesser accuracy by 2.08% and 12.86% in the GuessWhat?! benchmark, when no gold categories are available at inference time. The imagination module also boosts reasoning about object properties and attributes.</abstract>
<identifier type="citekey">suglia-etal-2020-imagining</identifier>
<identifier type="doi">10.18653/v1/2020.coling-main.95</identifier>
<location>
<url>https://aclanthology.org/2020.coling-main.95</url>
</location>
<part>
<date>2020-12</date>
<extent unit="page">
<start>1090</start>
<end>1102</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Imagining Grounded Conceptual Representations from Perceptual Information in Situated Guessing Games
%A Suglia, Alessandro
%A Vergari, Antonio
%A Konstas, Ioannis
%A Bisk, Yonatan
%A Bastianelli, Emanuele
%A Vanzo, Andrea
%A Lemon, Oliver
%Y Scott, Donia
%Y Bel, Nuria
%Y Zong, Chengqing
%S Proceedings of the 28th International Conference on Computational Linguistics
%D 2020
%8 December
%I International Committee on Computational Linguistics
%C Barcelona, Spain (Online)
%F suglia-etal-2020-imagining
%X In visual guessing games, a Guesser has to identify a target object in a scene by asking questions to an Oracle. An effective strategy for the players is to learn conceptual representations of objects that are both discriminative and expressive enough to ask questions and guess correctly. However, as shown by Suglia et al. (2020), existing models fail to learn truly multi-modal representations, relying instead on gold category labels for objects in the scene both at training and inference time. This provides an unnatural performance advantage when categories at inference time match those at training time, and it causes models to fail in more realistic “zero-shot” scenarios where out-of-domain object categories are involved. To overcome this issue, we introduce a novel “imagination” module based on Regularized Auto-Encoders, that learns context-aware and category-aware latent embeddings without relying on category labels at inference time. Our imagination module outperforms state-of-the-art competitors by 8.26% gameplay accuracy in the CompGuessWhat?! zero-shot scenario (Suglia et al., 2020), and it improves the Oracle and Guesser accuracy by 2.08% and 12.86% in the GuessWhat?! benchmark, when no gold categories are available at inference time. The imagination module also boosts reasoning about object properties and attributes.
%R 10.18653/v1/2020.coling-main.95
%U https://aclanthology.org/2020.coling-main.95
%U https://doi.org/10.18653/v1/2020.coling-main.95
%P 1090-1102
Markdown (Informal)
[Imagining Grounded Conceptual Representations from Perceptual Information in Situated Guessing Games](https://aclanthology.org/2020.coling-main.95) (Suglia et al., COLING 2020)
ACL