@article{cachewhatlaststokenretentionformemorybou,
  title = {Cache What Lasts: Token Retention for Memory-Bounded KV Cache in LLMs},
  author = {Ngoc Bui and Shubham Sharma and Simran Lamba and Saumitra Mishra and Rex Ying},
  year = {2025},
  eprint = {2512.03324},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2512.03324},
}