@article{attentionandcompressionisallyouneedforco,
  title = {Attention and Compression is all you need for Controllably Efficient Language Models},
  author = {Jatin Prakash and Aahlad Puli and Rajesh Ranganath},
  year = {2025},
  eprint = {2511.05313},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2511.05313},
}