[1]
Y. Zhao, H. Wu, and B. Xu, “Leveraging Attention to Effectively Compress Prompts for Long-Context LLMs”, AAAI, vol. 39, no. 24, pp. 26048–26056, Apr. 2025.