@inproceedings{song-etal-2024-robust,
title = "A Robust Dual-debiasing {VQA} Model based on Counterfactual Causal Effect",
author = "Song, Lingyun and
Yang, Chengkun and
Li, Xuanyu and
Shang, Xuequn",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2024",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-emnlp.245",
pages = "4242--4252",
abstract = "Traditional VQA models are inherently vulnerable to language bias, resulting in a significant performance drop when encountering out-of-distribution datasets. The conventional VQA models suffer from language bias that indicates a spurious correlation between textual questions and answers. Given the outstanding effectiveness of counterfactual causal inference in eliminating bias, we propose a model agnostic dual-debiasing framework based on Counterfactual Causal Effect (DCCE), which explicitly models two types of language bias(i.e., shortcut and distribution bias) by separate branches under the counterfactual inference framework. The effects of both types ofbias on answer prediction can be effectively mitigated by subtracting direct effect of textual questions on answers from total effect ofvisual questions on answers. Experimental results demonstrate that our proposed DCCE framework significantly reduces language biasand achieves state-of-the-art performance on the benchmark datasets without requiring additional augmented data. Our code is available inhttps://github.com/sxycyck/dcce.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="song-etal-2024-robust">
<titleInfo>
<title>A Robust Dual-debiasing VQA Model based on Counterfactual Causal Effect</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lingyun</namePart>
<namePart type="family">Song</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chengkun</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuanyu</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuequn</namePart>
<namePart type="family">Shang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yaser</namePart>
<namePart type="family">Al-Onaizan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohit</namePart>
<namePart type="family">Bansal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Miami, Florida, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Traditional VQA models are inherently vulnerable to language bias, resulting in a significant performance drop when encountering out-of-distribution datasets. The conventional VQA models suffer from language bias that indicates a spurious correlation between textual questions and answers. Given the outstanding effectiveness of counterfactual causal inference in eliminating bias, we propose a model agnostic dual-debiasing framework based on Counterfactual Causal Effect (DCCE), which explicitly models two types of language bias(i.e., shortcut and distribution bias) by separate branches under the counterfactual inference framework. The effects of both types ofbias on answer prediction can be effectively mitigated by subtracting direct effect of textual questions on answers from total effect ofvisual questions on answers. Experimental results demonstrate that our proposed DCCE framework significantly reduces language biasand achieves state-of-the-art performance on the benchmark datasets without requiring additional augmented data. Our code is available inhttps://github.com/sxycyck/dcce.</abstract>
<identifier type="citekey">song-etal-2024-robust</identifier>
<location>
<url>https://aclanthology.org/2024.findings-emnlp.245</url>
</location>
<part>
<date>2024-11</date>
<extent unit="page">
<start>4242</start>
<end>4252</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Robust Dual-debiasing VQA Model based on Counterfactual Causal Effect
%A Song, Lingyun
%A Yang, Chengkun
%A Li, Xuanyu
%A Shang, Xuequn
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Findings of the Association for Computational Linguistics: EMNLP 2024
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F song-etal-2024-robust
%X Traditional VQA models are inherently vulnerable to language bias, resulting in a significant performance drop when encountering out-of-distribution datasets. The conventional VQA models suffer from language bias that indicates a spurious correlation between textual questions and answers. Given the outstanding effectiveness of counterfactual causal inference in eliminating bias, we propose a model agnostic dual-debiasing framework based on Counterfactual Causal Effect (DCCE), which explicitly models two types of language bias(i.e., shortcut and distribution bias) by separate branches under the counterfactual inference framework. The effects of both types ofbias on answer prediction can be effectively mitigated by subtracting direct effect of textual questions on answers from total effect ofvisual questions on answers. Experimental results demonstrate that our proposed DCCE framework significantly reduces language biasand achieves state-of-the-art performance on the benchmark datasets without requiring additional augmented data. Our code is available inhttps://github.com/sxycyck/dcce.
%U https://aclanthology.org/2024.findings-emnlp.245
%P 4242-4252
Markdown (Informal)
[A Robust Dual-debiasing VQA Model based on Counterfactual Causal Effect](https://aclanthology.org/2024.findings-emnlp.245) (Song et al., Findings 2024)
ACL