@inproceedings{liu-ye-2019-strong,
title = "A Strong and Robust Baseline for Text-Image Matching",
author = "Liu, Fangyu and
Ye, Rongtian",
editor = "Alva-Manchego, Fernando and
Choi, Eunsol and
Khashabi, Daniel",
booktitle = "Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop",
month = jul,
year = "2019",
address = "Florence, Italy",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/P19-2023",
doi = "10.18653/v1/P19-2023",
pages = "169--176",
abstract = "We review the current schemes of text-image matching models and propose improvements for both training and inference. First, we empirically show limitations of two popular loss (sum and max-margin loss) widely used in training text-image embeddings and propose a trade-off: a kNN-margin loss which 1) utilizes information from hard negatives and 2) is robust to noise as all K-most hardest samples are taken into account, tolerating pseudo negatives and outliers. Second, we advocate the use of Inverted Softmax (IS) and Cross-modal Local Scaling (CSLS) during inference to mitigate the so-called hubness problem in high-dimensional embedding space, enhancing scores of all metrics by a large margin.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-ye-2019-strong">
<titleInfo>
<title>A Strong and Robust Baseline for Text-Image Matching</title>
</titleInfo>
<name type="personal">
<namePart type="given">Fangyu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rongtian</namePart>
<namePart type="family">Ye</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2019-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop</title>
</titleInfo>
<name type="personal">
<namePart type="given">Fernando</namePart>
<namePart type="family">Alva-Manchego</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eunsol</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Khashabi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Florence, Italy</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We review the current schemes of text-image matching models and propose improvements for both training and inference. First, we empirically show limitations of two popular loss (sum and max-margin loss) widely used in training text-image embeddings and propose a trade-off: a kNN-margin loss which 1) utilizes information from hard negatives and 2) is robust to noise as all K-most hardest samples are taken into account, tolerating pseudo negatives and outliers. Second, we advocate the use of Inverted Softmax (IS) and Cross-modal Local Scaling (CSLS) during inference to mitigate the so-called hubness problem in high-dimensional embedding space, enhancing scores of all metrics by a large margin.</abstract>
<identifier type="citekey">liu-ye-2019-strong</identifier>
<identifier type="doi">10.18653/v1/P19-2023</identifier>
<location>
<url>https://aclanthology.org/P19-2023</url>
</location>
<part>
<date>2019-07</date>
<extent unit="page">
<start>169</start>
<end>176</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Strong and Robust Baseline for Text-Image Matching
%A Liu, Fangyu
%A Ye, Rongtian
%Y Alva-Manchego, Fernando
%Y Choi, Eunsol
%Y Khashabi, Daniel
%S Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop
%D 2019
%8 July
%I Association for Computational Linguistics
%C Florence, Italy
%F liu-ye-2019-strong
%X We review the current schemes of text-image matching models and propose improvements for both training and inference. First, we empirically show limitations of two popular loss (sum and max-margin loss) widely used in training text-image embeddings and propose a trade-off: a kNN-margin loss which 1) utilizes information from hard negatives and 2) is robust to noise as all K-most hardest samples are taken into account, tolerating pseudo negatives and outliers. Second, we advocate the use of Inverted Softmax (IS) and Cross-modal Local Scaling (CSLS) during inference to mitigate the so-called hubness problem in high-dimensional embedding space, enhancing scores of all metrics by a large margin.
%R 10.18653/v1/P19-2023
%U https://aclanthology.org/P19-2023
%U https://doi.org/10.18653/v1/P19-2023
%P 169-176
Markdown (Informal)
[A Strong and Robust Baseline for Text-Image Matching](https://aclanthology.org/P19-2023) (Liu & Ye, ACL 2019)
ACL
- Fangyu Liu and Rongtian Ye. 2019. A Strong and Robust Baseline for Text-Image Matching. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: Student Research Workshop, pages 169–176, Florence, Italy. Association for Computational Linguistics.