Zeng, H. et al. 2026. Lethe: Layer- and Time-Adaptive KV Cache Pruning for Reasoning-Intensive LLM Serving. Proceedings of the AAAI Conference on Artificial Intelligence. 40, 33 (Mar. 2026), 28103–28112. DOI:https://doi.org/10.1609/aaai.v40i33.40036.