@article{heavytailedclassimbalanceandwhyadam, title = {Heavy-Tailed Class Imbalance and Why Adam Outperforms Gradient Descent on Language Models}, author = {Frederik Kunstner and Robin Yadav and Alan Milligan and Mark Schmidt and Alberto Bietti}, year = {2024}, eprint = {2402.19449}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2402.19449v2}, }