The field of Wi-Fi sensing and Distributed Fiber Optic Sensing (DFOS) is moving towards developing more robust and generalizable models. Researchers are exploring the use of meta-learning, graph-based methods, and federated learning to improve the accuracy and reliability of these systems. One of the key challenges in this field is the issue of domain shift, where models trained in one environment or setup fail to generalize to new environments or setups. To address this challenge, researchers are proposing novel frameworks and architectures that can adapt to new domains with minimal labeled data. Another important aspect of this field is the evaluation of model robustness to adversarial attacks, which is critical for the safe deployment of these systems in real-world environments. Noteworthy papers in this area include: Statistically-Guided Dual-Domain Meta-Learning with Adaptive Multi-Prototype Aggregation for Distributed Fiber Optic Sensing, which proposes a novel meta-learning framework for cross-deployment DFOS activity identification. Scale What Counts, Mask What Matters: Evaluating Foundation Models for Zero-Shot Cross-Domain Wi-Fi Sensing, which establishes scaling trends on Wi-Fi CSI sensing and shows that data scale and diversity are key to domain generalization. Graph-based 3D Human Pose Estimation using WiFi Signals, which presents a graph-based framework that explicitly models skeletal topology for WiFi-based 3D human pose estimation. Towards Trustworthy Wi-Fi Sensing: Systematic Evaluation of Deep Learning Model Robustness to Adversarial Attacks, which presents a systematic evaluation of the robustness of CSI deep learning models under diverse threat models. FedAPA: Federated Learning with Adaptive Prototype Aggregation Toward Heterogeneous Wi-Fi CSI-based Crowd Counting, which proposes a collaborative Wi-Fi CSI-based sensing algorithm that uses adaptive prototype aggregation strategy to assign similarity-based weights to peer prototypes.