@article{llmsknowwhattodropselfattentionguided,
  title = {LLMs Know What to Drop: Self-Attention Guided KV Cache Eviction for Efficient Long-Context Inference},
  author = {Guangtao Wang and Shubhangi Upasani and Chen Wu and Darshan Gandhi and Jonathan Li and Changran Hu and Bo Li and Urmish Thakker},
  year = {2025},
  eprint = {2503.08879},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2503.08879v1},
}