Zhan, J., Shen, H., Lin, Z., & He, T. (2026). PRISM: Privacy-Aware Routing for Adaptive Cloud–Edge LLM Inference via Semantic Sketch Collaboration. Proceedings of the AAAI Conference on Artificial Intelligence, 40(33), 28150–28158. https://doi.org/10.1609/aaai.v40i33.40041