Long-Context Generalization with Sparse Attention

@misc{martins2025longcontextgeneralizationsparse,
      title={Long-Context Generalization with Sparse Attention},
      author={André F. T. Martins and Pavlo Vasylenko and Marcos Treviso},
      year={2025},
      eprint={2506.16640},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2506.16640},
}