HyCal：面向跨学科少样本类增量学习的免训练原型校准方法
HyCal: A Training-Free Prototype Calibration Method for Cross-Discipline Few-Shot Class-Incremental Learning
在冻结的 CLIP 嵌入上，用余弦相似度与马氏距离互补地校准类原型，可作为一种免训练的原型修正手段，直接迁移到少样本/增量式的开放词表识别与跨域检索场景。
原文依据 a training-free method combining cosine similarity and Mahalanobis distance to capture complementary geometric properties
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论CLIP 等预训练视觉语言模型在持续学习中展现出潜力，但现有的少样本类增量学习方法假设数据来自同质领域且分布均衡，这限制了其在真实场景中的适用性——真实数据往往来自异质学科，样本可得性不均衡且视觉复杂度各异。我们识别出「领域引力」现象，即一种表征不对称性：异质领域间的数据不均衡使得过度代表或低熵领域对嵌入空间产生不成比例的影响，导致原型漂移，并使代表性不足或高熵领域上的性能下降。为解决该问题，我们提出跨学科可变少样本类增量学习（XD-VSCIL），这是一个刻画真实世界异质性与不均衡性的基准，领域引力在该设定下会自然加剧。我们提出混合原型校准（HyCal），一种免训练方法，结合余弦相似度与马氏距离以捕捉互补的几何性质——方向对齐与协方差感知的幅度——从而在不均衡异质条件下得到稳定的原型。HyCal 在冻结的 CLIP 嵌入上运行，在保持效率的同时持续提升保持与适应能力。实验表明，HyCal 有效缓解了领域引力，并在不均衡跨域增量学习中优于现有方法。
英文原文摘要 1438 字符
Pretrained Vision-Language Models (VLMs) like CLIP show promise in continual learning, but existing Few-Shot Class-Incremental Learning (FSCIL) methods assume homogeneous domains and balanced data distributions, limiting real-world applicability where data arises from heterogeneous disciplines with imbalanced sample availability and varying visual complexity. We identify Domain Gravity, a representational asymmetry where data imbalance across heterogeneous domains causes overrepresented or low-entropy domains to disproportionately influence the embedding space, leading to prototype drift and degraded performance on underrepresented or high-entropy domains. To address this, we introduce Cross-Discipline Variable Few-Shot Class-Incremental Learning (XD-VSCIL), a benchmark capturing real-world heterogeneity and imbalance where Domain Gravity naturally intensifies. We propose Hybrid Prototype Calibration (HyCal), a training-free method combining cosine similarity and Mahalanobis distance to capture complementary geometric properties--directional alignment and covariance-aware magnitude--yielding stable prototypes under imbalanced heterogeneous conditions. Operating on frozen CLIP embeddings, HyCal achieves consistent retention-adaptation improvements while maintaining efficiency. Experiments show HyCal effectively mitigates Domain Gravity and outperforms existing methods in imbalanced cross-domain incremental learning.