Federated Class-Incremental Learning (FCIL) is an underexplored yet pivotal issue, involving the dynamic addition of new classes in the context of federated learning. In this field, Data-Free Knowledge Transfer (DFKT) plays a crucial role in addressing catastrophic forgetting and data privacy problems. However, prior approaches lack the crucial synergy between DFKT and the model training phases, causing DFKT to encounter difficulties in generating high-quality data from a non-anchored latent space of the old task model. In this paper, we introduce LANDER (Label Text Centered Data-Free Knowledge Transfer) to address this issue by utilizing label text embeddings (LTE) produced by pretrained language models. Specifically, during the model training phase, our approach treats LTE as anchor points and constrains the feature embeddings of corresponding training samples around them, enriching the surrounding area with more meaningful information. In the DFKT phase, by using these LTE anchors, LANDER can synthesize more meaningful samples, thereby effectively addressing the forgetting problem. Additionally, instead of tightly constraining embeddings toward the anchor, the Bounding Loss is introduced to encourage sample embeddings to remain flexible within a defined radius. This approach preserves the natural differences in sample embeddings and mitigates the embedding overlap caused by heterogeneous federated settings. Extensive experiments conducted on CIFAR100, Tiny-ImageNet, and ImageNet demonstrate that LANDER significantly outperforms previous methods and achieves state-of-the-art performance in FCIL. The code is available at https://github.com/tmtuan1307/lander.

翻译：联邦类增量学习（FCIL）是一个尚未充分探索但至关重要的问题，涉及在联邦学习背景下动态添加新类别。在该领域，无数据知识迁移（DFKT）在解决灾难性遗忘和数据隐私问题中扮演关键角色。然而，现有方法缺乏DFKT与模型训练阶段之间的关键协同，导致DFKT难以从旧任务模型的非锚定潜在空间中生成高质量数据。本文提出LANDER（标签文本中心的无数据知识迁移），通过利用预训练语言模型生成的标签文本嵌入（LTE）来解决该问题。具体而言，在模型训练阶段，我们的方法将LTE作为锚定点，约束对应训练样本的特征嵌入围绕其分布，从而在锚点周围区域丰富更有意义的信息。在DFKT阶段，通过利用这些LTE锚点，LANDER能够合成更具代表性的样本，有效缓解遗忘问题。此外，为替代对嵌入的紧密锚定约束，我们引入边界损失（Bounding Loss），鼓励样本嵌入在定义的半径范围内保持灵活性。该方法保留了样本嵌入的自然差异性，并缓解了异构联邦设置导致的嵌入重叠问题。在CIFAR100、Tiny-ImageNet和ImageNet上的大量实验表明，LANDER显著优于以往方法，并在FCIL中达到最优性能。代码开源地址：https://github.com/tmtuan1307/lander。