This paper addresses the critical data scarcity that hinders the practical deployment of learning to defer (L2D) systems to the population. We introduce a context-aware, semi-supervised framework that uses meta-learning to generate expert-specific embeddings from only a few demonstrations. We demonstrate the efficacy of a dual-purpose mechanism, where these embeddings are used first to generate a large corpus of pseudo-labels for training, and subsequently to enable on-the-fly adaptation to new experts at test-time. The experiment results on three different datasets confirm that a model trained on these synthetic labels rapidly approaches oracle-level performance, validating the data efficiency of our approach. By resolving a key training bottleneck, this work makes adaptive L2D systems more practical and scalable, paving the way for human-AI collaboration in real-world environments. To facilitate reproducibility and address implementation details not covered in the main text, we provide our source code and training configurations at https://github.com/nil123532/learning-to-defer-to-a-population-with-limited-demonstrations.

翻译：本文针对学习推诿（L2D）系统在实际部署到群体时面临的关键数据稀缺问题展开研究。我们提出了一种情境感知的半监督框架，该框架利用元学习技术，仅通过少量示范即可生成专家特定的嵌入表示。我们论证了一种双重用途机制的有效性：这些嵌入首先用于生成大量伪标签进行训练，随后在测试阶段实现对新专家的即时适应。在三个不同数据集上的实验结果表明，基于这些合成标签训练的模型能快速逼近理论最优性能，验证了本方法的数据高效性。通过解决关键的训练瓶颈，本研究使得自适应L2D系统更具实用性和可扩展性，为现实环境中的人机协作铺平道路。为促进可复现性并补充正文未涵盖的实现细节，我们在 https://github.com/nil123532/learning-to-defer-to-a-population-with-limited-demonstrations 公开了源代码与训练配置。