Cause-Effect Driven Optimization for Robust Medical Visual Question Answering with Language Biases
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915354792427520 |
|---|---|
| author | Zhu, Huanjia Liu, Yishu Fang, Xiaozhao Lu, Guangming Chen, Bingzhi |
| author_facet | Zhu, Huanjia Liu, Yishu Fang, Xiaozhao Lu, Guangming Chen, Bingzhi |
| contents | Existing Medical Visual Question Answering (Med-VQA) models often suffer from language biases, where spurious correlations between question types and answer categories are inadvertently established. To address these issues, we propose a novel Cause-Effect Driven Optimization framework called CEDO, that incorporates three well-established mechanisms, i.e., Modality-driven Heterogeneous Optimization (MHO), Gradient-guided Modality Synergy (GMS), and Distribution-adapted Loss Rescaling (DLR), for comprehensively mitigating language biases from both causal and effectual perspectives. Specifically, MHO employs adaptive learning rates for specific modalities to achieve heterogeneous optimization, thus enhancing robust reasoning capabilities. Additionally, GMS leverages the Pareto optimization method to foster synergistic interactions between modalities and enforce gradient orthogonality to eliminate bias updates, thereby mitigating language biases from the effect side, i.e., shortcut bias. Furthermore, DLR is designed to assign adaptive weights to individual losses to ensure balanced learning across all answer categories, effectively alleviating language biases from the cause side, i.e., imbalance biases within datasets. Extensive experiments on multiple traditional and bias-sensitive benchmarks consistently demonstrate the robustness of CEDO over state-of-the-art competitors. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_17903 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Cause-Effect Driven Optimization for Robust Medical Visual Question Answering with Language Biases Zhu, Huanjia Liu, Yishu Fang, Xiaozhao Lu, Guangming Chen, Bingzhi Computer Vision and Pattern Recognition Artificial Intelligence Existing Medical Visual Question Answering (Med-VQA) models often suffer from language biases, where spurious correlations between question types and answer categories are inadvertently established. To address these issues, we propose a novel Cause-Effect Driven Optimization framework called CEDO, that incorporates three well-established mechanisms, i.e., Modality-driven Heterogeneous Optimization (MHO), Gradient-guided Modality Synergy (GMS), and Distribution-adapted Loss Rescaling (DLR), for comprehensively mitigating language biases from both causal and effectual perspectives. Specifically, MHO employs adaptive learning rates for specific modalities to achieve heterogeneous optimization, thus enhancing robust reasoning capabilities. Additionally, GMS leverages the Pareto optimization method to foster synergistic interactions between modalities and enforce gradient orthogonality to eliminate bias updates, thereby mitigating language biases from the effect side, i.e., shortcut bias. Furthermore, DLR is designed to assign adaptive weights to individual losses to ensure balanced learning across all answer categories, effectively alleviating language biases from the cause side, i.e., imbalance biases within datasets. Extensive experiments on multiple traditional and bias-sensitive benchmarks consistently demonstrate the robustness of CEDO over state-of-the-art competitors. |
| title | Cause-Effect Driven Optimization for Robust Medical Visual Question Answering with Language Biases |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2506.17903 |