让医学视觉问答模型自动剔除数据偏见，专注真实诊断证据。
Learning to Trim: End-to-End Causal Graph Pruning with Dynamic Anatomical Feature Banks for Medical VQA
- 用动态特征库捕捉常见解剖模式，作为数据偏差的代理
- 可微剪枝模块软抑制与全局模式强相关的特征，突出个体证据
- 在多个医学VQA数据集上显著提升模型泛化能力
医学视觉问答（MedVQA）模型常因依赖数据集特有的关联（如重复的解剖结构或问题类型规律）而泛化能力有限，而非真正诊断依据。现有因果方法多为静态调整或事后修正。为此，我们提出可学习的因果剪枝（LCT）框架，将因果剪枝融入端到端优化。引入动态解剖特征库（DAFB），通过动量机制更新，捕捉频繁出现的解剖与语言模式的全局原型，近似数据级规律。设计可微剪枝模块，估计实例表示与全局特征库间的依赖关系：与全局原型高度相关的特征被软抑制，而实例特异性证据被强化。该可学习机制使模型能自适应地优先关注因果信号而非虚假相关。在VQA-RAD、SLAKE、SLAKE-CP和PathVQA上的实验表明，LCT在鲁棒性和泛化性上持续优于现有去偏策略。
原文摘要 · Abstract (English)
Medical Visual Question Answering (MedVQA) models often exhibit limited generalization due to reliance on dataset-specific correlations, such as recurring anatomical patterns or question-type regularities, rather than genuine diagnostic evidence. Existing causal approaches are typically implemented as static adjustments or post-hoc corrections. To address this issue, we propose a Learnable Causal Trimming (LCT) framework that integrates causal pruning into end-to-end optimization. We introduce a Dynamic Anatomical Feature Bank (DAFB), updated via a momentum mechanism, to capture global prototypes of frequent anatomical and linguistic patterns, serving as an approximation of dataset-level regularities. We further design a differentiable trimming module that estimates the dependency between instance-level representations and the global feature bank. Features highly correlated with global prototypes are softly suppressed, while instance-specific evidence is emphasized. This learnable mechanism encourages the model to prioritize causal signals over spurious correlations adaptively. Experiments on VQA-RAD, SLAKE, SLAKE-CP and PathVQA demonstrate that LCT consistently improves robustness and generalization over existing debiasing strategies.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。