A Robust Dual-debiasing VQA Model based on Counterfactual Causal Effect

Lingyun Song, Chengkun Yang, Xuanyu Li, Xuequn Shang


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.
Anthology ID:
2024.findings-emnlp.245
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2024
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4242–4252
Language:
URL:
https://aclanthology.org/2024.findings-emnlp.245
DOI:
Bibkey:
Cite (ACL):
Lingyun Song, Chengkun Yang, Xuanyu Li, and Xuequn Shang. 2024. A Robust Dual-debiasing VQA Model based on Counterfactual Causal Effect. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 4242–4252, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
A Robust Dual-debiasing VQA Model based on Counterfactual Causal Effect (Song et al., Findings 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.findings-emnlp.245.pdf