Compositional Generalization in Image Captioning

Mitja Nikolaus; Mostafa Abdou; Matthew Lamm; Rahul Aralikatte; Desmond Elliott

doi:10.18653/v1/k19-1009

Compositional Generalization in Image Captioning

Mitja Nikolaus, Mostafa Abdou, Matthew Lamm, Rahul Aralikatte, Desmond Elliott

4 Citations (Scopus)

Abstract

Image captioning models are usually evaluated on their ability to describe a held-out set of images, not on their ability to generalize to unseen concepts. We study the problem of compositional generalization, which measures how well a model composes unseen combinations of concepts when describing images. State-of-the-art image captioning models show poor generalization performance on this task. We propose a multi-task model to address the poor performance, that combines caption generation and image--sentence ranking, and uses a decoding mechanism that re-ranks the captions according their similarity to the image. This model is substantially better at generalizing to unseen combinations of concepts compared to state-of-the-art captioning models.

Original language	Undefined/Unknown
Title of host publication	Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL)
Number of pages	12
Place of Publication	Hong Kong, China
Publisher	Association for Computational Linguistics (ACL)
Publication date	1 Nov 2019
Pages	87-98
DOIs	https://doi.org/10.18653/v1/k19-1009
Publication status	Published - 1 Nov 2019

Access to Document

10.18653/v1/k19-1009

K19-1009.pdfFinal published version, 807 KBLicence: CC BY

https://www.aclweb.org/anthology/K19-1009

Cite this

Compositional Generalization in Image Captioning. / Nikolaus, Mitja; Abdou, Mostafa; Lamm, Matthew et al.
Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL). Hong Kong, China: Association for Computational Linguistics (ACL), 2019. p. 87-98.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

@inproceedings{2f18fd294bcb4b66b1fa9decac2c8b6c,

title = "Compositional Generalization in Image Captioning",

abstract = "Image captioning models are usually evaluated on their ability to describe a held-out set of images, not on their ability to generalize to unseen concepts. We study the problem of compositional generalization, which measures how well a model composes unseen combinations of concepts when describing images. State-of-the-art image captioning models show poor generalization performance on this task. We propose a multi-task model to address the poor performance, that combines caption generation and image--sentence ranking, and uses a decoding mechanism that re-ranks the captions according their similarity to the image. This model is substantially better at generalizing to unseen combinations of concepts compared to state-of-the-art captioning models.",

author = "Mitja Nikolaus and Mostafa Abdou and Matthew Lamm and Rahul Aralikatte and Desmond Elliott",

year = "2019",

month = nov,

day = "1",

doi = "10.18653/v1/k19-1009",

language = "Udefineret/Ukendt",

pages = "87--98",

booktitle = "Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL)",

publisher = "Association for Computational Linguistics (ACL)",

address = "USA",

}

TY - GEN

T1 - Compositional Generalization in Image Captioning

AU - Nikolaus, Mitja

AU - Abdou, Mostafa

AU - Lamm, Matthew

AU - Aralikatte, Rahul

AU - Elliott, Desmond

PY - 2019/11/1

Y1 - 2019/11/1

N2 - Image captioning models are usually evaluated on their ability to describe a held-out set of images, not on their ability to generalize to unseen concepts. We study the problem of compositional generalization, which measures how well a model composes unseen combinations of concepts when describing images. State-of-the-art image captioning models show poor generalization performance on this task. We propose a multi-task model to address the poor performance, that combines caption generation and image--sentence ranking, and uses a decoding mechanism that re-ranks the captions according their similarity to the image. This model is substantially better at generalizing to unseen combinations of concepts compared to state-of-the-art captioning models.

AB - Image captioning models are usually evaluated on their ability to describe a held-out set of images, not on their ability to generalize to unseen concepts. We study the problem of compositional generalization, which measures how well a model composes unseen combinations of concepts when describing images. State-of-the-art image captioning models show poor generalization performance on this task. We propose a multi-task model to address the poor performance, that combines caption generation and image--sentence ranking, and uses a decoding mechanism that re-ranks the captions according their similarity to the image. This model is substantially better at generalizing to unseen combinations of concepts compared to state-of-the-art captioning models.

U2 - 10.18653/v1/k19-1009

DO - 10.18653/v1/k19-1009

M3 - Konferencebidrag i proceedings

SP - 87

EP - 98

BT - Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL)

PB - Association for Computational Linguistics (ACL)

CY - Hong Kong, China

ER -