@article{bypassingtheexponentialdependencylooped, title = {Bypassing the Exponential Dependency: Looped Transformers Efficiently Learn In-context by Multi-step Gradient Descent}, author = {Bo Chen and Xiaoyu Li and YIngyu Liang and Zhenmei Shi and Zhao Song}, year = {2024}, eprint = {2410.11268}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2410.11268v1}, }