摘要：现代视觉问答（VQA）模型常会利用训练数据中的伪相关性，因语言偏见而导致较差的分布外（OOD）泛化能力。尽管反事实学习已显示出前景，现有方法仍可进一步改进，以更好地引导注意力聚焦于因果证据，并增强特征区分能力。为解决这一问题，我们提出了一种新颖的训练框架，用于增强 VQA 的反事实对比学习。我们的框架包含三项关键贡献：（1）用于实现稳定多目标优化的三阶段课程学习；（2）用于学习更具区分性特征的增强型 Batch-Contrastive 损失；以及（3）两种新颖的正则化方法：用于细化预测空间的 Answer-Contrastive（AC）损失，以及用于确保因果视觉落地的 Gradient-Discrepancy（GD）损失。我们的模型在偏见敏感的 VQA-CP v2 基准上取得了具有竞争力的 61.64% 准确率，同时在标准 VQA v2 数据集上保持 62.80%，泛化差距仅为 1.16%。这表明该方法在 OOD 鲁棒性与分布内性能之间实现了良好平衡。

Abstract:Modern Visual Question Answering (VQA) models often exploit spurious correlations in training data, leading to poor out-of-distribution (OOD) generalization due to language bias. Although counterfactual learning has shown promise, existing methods can be improved to better guide attention toward causal evidence and strengthen feature discrimination. To address this, we propose a novel training framework that enhances counterfactual contrastive learning for VQA. Our framework introduces three key contributions: (1) a three-stage curriculum for stable multi-objective optimization, (2) an enhanced Batch-Contrastive loss for more discriminative feature learning, and (3) two novel regularizers, Answer-Contrastive (AC) loss to refine the prediction space and Gradient-Discrepancy (GD) loss to enforce causal visual grounding. Our model achieves a competitive accuracy of 61.64% on the bias-sensitive VQA-CP v2 benchmark while maintaining 62.80% on the standard VQA v2 dataset, yielding a small generalization gap of 1.16%. This demonstrates a strong balance between OOD robustness and in-distribution performance.