@article{recallbeforeyouranksimilarityguidedtopkr, title = {Recall Before You Rank: Similarity-Guided Top-$K$ Reuse for Efficient Long-Context Attention}, author = {Wenshuai Yao and Wenyong Zhou and Hanyong Shao and Yizhe Chen and Zhiyuan Ning and Yuannuo Feng and Ru Huang and Kechao Tang}, year = {2026}, eprint = {2607.27692}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2607.27692}, }