@article{kascadeapracticalsparseattentionmethodfo, title = {Kascade: A Practical Sparse Attention Method for Long-Context LLM Inference}, author = {Dhruv Deshmukh and Saurabh Goyal and Nipun Kwatra and Ramachandran Ramjee}, year = {2025}, eprint = {2512.16391}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2512.16391}, }