The evolution of next-generation wireless networks demands intelligent, adaptive, and explainable decision-making for robust communication in dynamic environments. This paper presentsEnwar 2.0, the first agentic large language model (LLM) framework integrating adaptive retrieval-augmented generation (RAG) and chain-of-thought (CoT) reasoning into situation-aware and explainable wireless network management.Enwar 2.0introduces two specialized agents: a transformer-fusion (TransFusion)-based beam prediction agent and an environment perception agent, both of which fuse multi-modal sensory inputs—including camera, LiDAR, radar, and GPS—from the DeepSense6G dataset. The beam prediction agent enables infrastructure-to-vehicle (I2V) target-in-the-loop beam tracking and real-time adaptation based on dynamic environmental conditions. In contrast, the environment perception agent provides situation-aware reasoning and justifications for beam decisions. Unlike its predecessor,Enwar 1.0, which relied on static knowledge bases (KBs) and text-only LLMs,Enwar 2.0is designed for CoT reasoning, leverages LLaMa3.2-3B/LLaMa3.1-8B/LLaMa3.3-70B for text-generation, the multi-modal capabilities of LLaMa 3.2, and employs LlamaIndex for fine-grained, dynamic context retrieval, eliminating retrieval ambiguities and enhancing response relevance. Numerical results show that the beam prediction agent achieves up to 90.0% Top-3 accuracy at$t+3$, effectively predicting optimal beam selections three time steps ahead. Overall,Enwar 2.0achieves state-of-the-art performance, with up to 89.7%/83.5% interpretation/perception correctness, 81.6%/80.9% faithfulness, and 89.9%/88.2% relevancy. In comparison, the baseline pretrained LLaMa3 models without adaptive RAG achieves up to 80.3%/77.3% correctness, and the baseline without RAG performs significantly worse at 67.1%/64.8%. Additionally,Enwar 2.0reduces processing time by over 100% relative to the baseline, while its adaptive RAG improves performance by up to 13.7% compared to static RAG.

Date of Conference: 04-07 August 2025. Conference Location: Tokyo, Japan

Team-Based Learning (TBL), a pedagogical approach that positively influences classroom attendance, still needs help with student absenteeism. Current attendance tracking tools are designed for something other than TBL environments and require manual interaction from the instructor or students, consuming valuable class time. This paper introduces a novel approach to this problem, proposing an automated attendance tool for TBL classes using fixed sensors, either a camera or a Light Detection and Ranging (LiDAR), combined with a machine learning classification algorithm. The paper delves into a comparative study of using cameras and LiDAR for this purpose, evaluating them based on privacy, accuracy, efficiency, and perceptions of students and instructors. The results indicate that while both methods successfully record attendance, the LiDAR system was more efficient and reliable. Although the camera offered a higher accuracy rate, a better customized LiDAR dataset designed for the classroom environment could enhance the machine learning algorithm's accuracy in identifying students and recording attendance. Finally, the LiDAR system was favored by both students and instructors for its ease of use, non-intrusiveness, and privacy preservation.