Sanae Lotfi

~Sanae_Lotfi1

4

论文总数

2.0

年均投稿

6.7

平均评分

接收情况3/4

会议分布

NeurIPS

2

ICLR

1

ICML

1

发表论文 (4 篇)

20252 篇

Small Batch Size Training for Language Models: When Vanilla SGD Works, and Why Gradient Accumulation is Wasteful

NeurIPS 2025Poster

Customizing the Inductive Biases of Softmax Attention using Structured Matrices

ICML 2025Poster

20242 篇

Unlocking Tokens as Data Points for Generalization Bounds on Larger Language Models

NeurIPS 2024Spotlight

Non-Vacuous Generalization Bounds for Large Language Models

ICLR 2024Rejected

合作者 (11)

Andrew Gordon Wilson4 篇

Micah Goldblum3 篇

Yilun Kuang3 篇

Marc Anton Finzi2 篇

Aditya Somasundaram1 篇

Martin Marek1 篇

Brandon Amos1 篇

Tim G. J. Rudner1 篇