BayesVQA: Energy-Guided Bayesian Debiasing for Language-Bias-Robust Visual Question Answering

Zhiqi Huang; Huanjia Zhu; Xiangwen Deng; Zhong Qinghao; Bingzhi Chen

AAAI 2026technical0 citations

BayesVQA: Energy-Guided Bayesian Debiasing for Language-Bias-Robust Visual Question Answering

Zhiqi Huang, Huanjia Zhu, Xiangwen Deng, Zhong Qinghao, Bingzhi Chen

Abstract

Numerous studies have demonstrated that Visual Question Answering (VQA) models are vulnerable to language priors and dataset biases, often leading to spurious correlations between questions and answers. As a result, these models excessively rely on linguistic cues, neglecting essential visual information and causing representational distortions. To address this challenge, we propose a novel Bayesian debiasing framework termed BayesVQA, which integrates three carefully designed mechanisms: Energy-guided Prior Variance (EPV), Energy-guided Posterior Sampling (EPS), and Energy-guided Likelihood Reweighting (ELR). Specifically, we explicitly decompose each sample

BibTeX

@inproceedings{aaai2026_bayesvqaenergygu,
  title = {BayesVQA: Energy-Guided Bayesian Debiasing for Language-Bias-Robust Visual Question Answering},
  author = {Zhiqi Huang and Huanjia Zhu and Xiangwen Deng and Zhong Qinghao and Bingzhi Chen},
  booktitle = {AAAI 2026},
  year = {2026}
}