Zhizhen Li

Zhizhen Li, Xuanhao Luo, Xueren Ge et al.

Large AI models have been widely adopted in wireless communications for channel modeling, beamforming, and resource optimization. However, most existing efforts remain limited to single-modality inputs and channel-specific objec- tives, overlooking the broader potential of large foundation models for unified wireless sensing. To bridge this gap, we propose MMSense, a multi-modal, multi-task foundation model that jointly addresses channel-centric, environment-aware, and human-centered sensing. Our framework integrates image, radar, LiDAR, and textual data by transforming them into vision- compatible representations, enabling effective cross-modal align- ment within a unified feature space. A modality gating mecha- nism adaptively fuses these representations, while a vision-based large language model backbone enables unified feature align- ment and instruction-driven task adaptation. Furthermore, task- specific sequential attention and uncertainty-based loss weighting mechanisms enhance cross-task generalization. Experiments on real wireless scenario datasets show that our approach outper- forms both task-specific and large-model baselines, confirming its strong generalization across heterogeneous sensing tasks.

Xuanhao Luo, Zhizhen Li, Yuchen Liu

Channel state information (CSI) provides a widely available sensing modality for human and environment perception, but existing CSI sensing models usually rely on task-specific supervised training and require substantial labeled data for each task, device, user, or environment. This limits their scalability in practical deployments where unlabeled CSI is abundant but labeled data is costly to collect. In this paper, we present CSI-JEPA, a self-supervised predictive representation learning framework for label-efficient, multi-task Wi-Fi sensing. CSI-JEPA learns reusable temporal-spectral representations from unlabeled CSI samples by predicting latent features of masked channel regions from visible context. To better match the physical structure of CSI, CSI-JEPA tokenizes channel-response amplitude windows along the time and subcarrier dimensions. It then introduces a channel variation-aware masking strategy that samples predictive targets from regions with stronger local temporal and subcarrier-domain variations. After pretraining, the encoder is frozen and used as a backbone, with lightweight task-specific adapters added for downstream sensing tasks. We evaluate CSI-JEPA on seven real-world Wi-Fi sensing tasks spanning diverse objectives and deployment settings. The results show that CSI-JEPA improves downstream sensing performance over competitive baselines, achieving up to 10.64 percentage points mean accuracy gain over state-of-the-art supervised Transformer and matched-budget label savings of up to 98.0%.

zhizhen li, tianyi zhuo, Yifei Cao et al.

Video stabilization often struggles with distortion and excessive cropping. This paper proposes a novel end-to-end framework, named TranStable, to address these challenges, comprising a genera tor and a discriminator. We establish TransformerUNet (TUNet) as the generator to utilize the Hierarchical Adaptive Fusion Module (HAFM), integrating Transformer and CNN to leverage both global and local features across multiple visual cues. By modeling frame-wise relationships, it generates robust pixel-level warping maps for stable geometric transformations. Furthermore, we design the Stability Discriminator Module (SDM), which provides pixel-wise supervision for authenticity and consistency in training period, ensuring more complete field-of-view while minimizing jitter artifacts and enhancing visual fidelity. Extensive experiments on NUS, DeepStab, and Selfie benchmarks demonstrate state-of-the-art performance.

Xuanhao Luo, Shivesh Madan Nath Jha, Akruti Sinha et al.

Network log data analysis plays a critical role in detecting security threats and operational anomalies. Traditional log analysis methods for anomaly detection and root cause analysis rely heavily on expert knowledge or fully supervised learning models, both of which require extensive labeled data and significant human effort. To address these challenges, we propose ALPHA, the first Active Learning Pipeline for Human-free log Analysis. ALPHA integrates semantic embedding, clustering-based representative sampling, and large language model (LLM)-assisted few-shot annotation to automate the anomaly detection process. The LLM annotated labels are propagated across clusters, enabling large-scale training of an anomaly detector with minimal supervision. To enhance the annotation accuracy, we propose a two-step few-shot refinement strategy that adaptively selects informative prompts based on the LLM's observed error patterns. Extensive experiments on real-world log datasets demonstrate that ALPHA achieves detection accuracy comparable to fully supervised methods while mitigating human efforts in the loop. ALPHA also supports interpretable analysis through LLM-driven root cause explanations in the post-detection stage. These capabilities make ALPHA a scalable and cost-efficient solution for truly automated log-based anomaly detection.