Lotfi, Sanae

10 publications

ICML 2025 Customizing the Inductive Biases of SoftMax Attention Using Structured Matrices Yilun Kuang, Noah Amsel, Sanae Lotfi, Shikai Qiu, Andres Potapczynski, Andrew Gordon Wilson

NeurIPS 2025 Small Batch Size Training for Language Models: When Vanilla SGD Works, and Why Gradient Accumulation Is Wasteful Martin Marek, Sanae Lotfi, Aditya Somasundaram, Andrew Gordon Wilson, Micah Goldblum

ICML 2024 Non-Vacuous Generalization Bounds for Large Language Models Sanae Lotfi, Marc Anton Finzi, Yilun Kuang, Tim G. J. Rudner, Micah Goldblum, Andrew Gordon Wilson

NeurIPS 2024 Unlocking Tokens as Data Points for Generalization Bounds on Larger Language Models Sanae Lotfi, Yilun Kuang, Brandon Amos, Micah Goldblum, Marc Finzi, Andrew Gordon Wilson

ICMLW 2024 Unlocking Tokens as Data Points for Generalization Bounds on Larger Language Models Sanae Lotfi, Yilun Kuang, Marc Anton Finzi, Brandon Amos, Micah Goldblum, Andrew Gordon Wilson

NeurIPSW 2023 Non-Vacuous Generalization Bounds for Large Language Models Sanae Lotfi, Marc Finzi, Yilun Kuang, Tim Rudner, Micah Goldblum, Andrew Wilson

ICML 2022 Bayesian Model Selection, the Marginal Likelihood, and Generalization Sanae Lotfi, Pavel Izmailov, Gregory Benton, Micah Goldblum, Andrew Gordon Wilson

NeurIPS 2022 PAC-Bayes Compression Bounds so Tight That They Can Explain Generalization Sanae Lotfi, Marc Finzi, Sanyam Kapoor, Andres Potapczynski, Micah Goldblum, Andrew G Wilson

NeurIPS 2021 Dangers of Bayesian Model Averaging Under Covariate Shift Pavel Izmailov, Patrick Nicholson, Sanae Lotfi, Andrew G Wilson

ICML 2021 Loss Surface Simplexes for Mode Connecting Volumes and Fast Ensembling Gregory Benton, Wesley Maddox, Sanae Lotfi, Andrew Gordon Gordon Wilson