@inproceedings{melas-kyriazi-etal-2018-training,
title = "Training for Diversity in Image Paragraph Captioning",
author = "Melas-Kyriazi, Luke and
Rush, Alexander and
Han, George",
editor = "Riloff, Ellen and
Chiang, David and
Hockenmaier, Julia and
Tsujii, Jun{'}ichi",
booktitle = "Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing",
month = oct # "-" # nov,
year = "2018",
address = "Brussels, Belgium",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/D18-1084",
doi = "10.18653/v1/D18-1084",
pages = "757--761",
abstract = "Image paragraph captioning models aim to produce detailed descriptions of a source image. These models use similar techniques as standard image captioning models, but they have encountered issues in text generation, notably a lack of diversity between sentences, that have limited their effectiveness. In this work, we consider applying sequence-level training for this task. We find that standard self-critical training produces poor results, but when combined with an integrated penalty on trigram repetition produces much more diverse paragraphs. This simple training approach improves on the best result on the Visual Genome paragraph captioning dataset from 16.9 to 30.6 CIDEr, with gains on METEOR and BLEU as well, without requiring any architectural changes.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="melas-kyriazi-etal-2018-training">
<titleInfo>
<title>Training for Diversity in Image Paragraph Captioning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luke</namePart>
<namePart type="family">Melas-Kyriazi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Rush</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">George</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2018-oct-nov</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ellen</namePart>
<namePart type="family">Riloff</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Chiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Julia</namePart>
<namePart type="family">Hockenmaier</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun’ichi</namePart>
<namePart type="family">Tsujii</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Brussels, Belgium</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Image paragraph captioning models aim to produce detailed descriptions of a source image. These models use similar techniques as standard image captioning models, but they have encountered issues in text generation, notably a lack of diversity between sentences, that have limited their effectiveness. In this work, we consider applying sequence-level training for this task. We find that standard self-critical training produces poor results, but when combined with an integrated penalty on trigram repetition produces much more diverse paragraphs. This simple training approach improves on the best result on the Visual Genome paragraph captioning dataset from 16.9 to 30.6 CIDEr, with gains on METEOR and BLEU as well, without requiring any architectural changes.</abstract>
<identifier type="citekey">melas-kyriazi-etal-2018-training</identifier>
<identifier type="doi">10.18653/v1/D18-1084</identifier>
<location>
<url>https://aclanthology.org/D18-1084</url>
</location>
<part>
<date>2018-oct-nov</date>
<extent unit="page">
<start>757</start>
<end>761</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Training for Diversity in Image Paragraph Captioning
%A Melas-Kyriazi, Luke
%A Rush, Alexander
%A Han, George
%Y Riloff, Ellen
%Y Chiang, David
%Y Hockenmaier, Julia
%Y Tsujii, Jun’ichi
%S Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing
%D 2018
%8 oct nov
%I Association for Computational Linguistics
%C Brussels, Belgium
%F melas-kyriazi-etal-2018-training
%X Image paragraph captioning models aim to produce detailed descriptions of a source image. These models use similar techniques as standard image captioning models, but they have encountered issues in text generation, notably a lack of diversity between sentences, that have limited their effectiveness. In this work, we consider applying sequence-level training for this task. We find that standard self-critical training produces poor results, but when combined with an integrated penalty on trigram repetition produces much more diverse paragraphs. This simple training approach improves on the best result on the Visual Genome paragraph captioning dataset from 16.9 to 30.6 CIDEr, with gains on METEOR and BLEU as well, without requiring any architectural changes.
%R 10.18653/v1/D18-1084
%U https://aclanthology.org/D18-1084
%U https://doi.org/10.18653/v1/D18-1084
%P 757-761
Markdown (Informal)
[Training for Diversity in Image Paragraph Captioning](https://aclanthology.org/D18-1084) (Melas-Kyriazi et al., EMNLP 2018)
ACL
- Luke Melas-Kyriazi, Alexander Rush, and George Han. 2018. Training for Diversity in Image Paragraph Captioning. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 757–761, Brussels, Belgium. Association for Computational Linguistics.