@inproceedings{vccscalingtransformersto128ktokensor, title = {Vcc: Scaling Transformers to 128K Tokens or More by Prioritizing Important Tokens}, year = {2023}, booktitle = {NeurIPS 2023 11}, eprint = {2305.04241}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2305.04241v2}, }