@article{criticalbatchsizerevisitedasimple, title = {Critical Batch Size Revisited: A Simple Empirical Approach to Large-Batch Language Model Training}, author = {William Merrill and Shane Arora and Dirk Groeneveld and Hannaneh Hajishirzi}, year = {2025}, eprint = {2505.23971}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2505.23971v1}, }