@article{overfilltwostagemodelsforefficientlangua, title = {OverFill: Two-Stage Models for Efficient Language Model Decoding}, author = {Woojeong Kim and Junxiong Wang and Jing Nathan Yan and Mohamed Abdelfattah and Alexander M. Rush}, year = {2025}, eprint = {2508.08446}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2508.08446}, }