The field of collaborative perception is moving towards developing more efficient and effective communication strategies for multi-agent systems. Researchers are exploring innovative approaches to optimize the trade-off between task performance and communication volume, leveraging information theory and semantic-aware paradigms. Notable advancements include the development of rate-distortion optimized communication frameworks, instance-level interaction architectures, and prediction-powered communication methods with distortion guarantees. These innovations have the potential to significantly improve the accuracy and robustness of collaborative perception systems in various applications, including autonomous vehicles and edge inference. Noteworthy papers include: Rate-Distortion Optimized Communication for Collaborative Perception, which introduces a pragmatic rate-distortion theory for multi-agent collaboration and proposes a communication-efficient framework called RDcomm. INSTINCT: Instance-Level Interaction Architecture for Query-Based Collaborative Perception, which presents a novel collaborative perception framework featuring a quality-aware filtering mechanism and a dual-branch detection routing scheme. Prediction-Powered Communication with Distortion Guarantees, which proposes two zero-delay compression algorithms leveraging online conformal prediction to provide per-sequence guarantees on the distortion of reconstructed sequences.