Lune

Profiles2020-2023

Behnam Neyshabur

Carnegie Mellon University

16Papers
8Last author
29Unique co-authors
1First author
1Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers16

  1. Long Range Language Modeling via Gated State SpacesAuthor 4 of 4ICLR 2023 · 45 citations
  2. Exploring Length Generalization in Large Language ModelsAuthor 10 of 10NeurIPS 2022 · 267 citations
  3. Revisiting Neural Scaling Laws in Language and VisionAuthor 2 of 3NeurIPS 2022 · 171 citations
  4. Block-Recurrent TransformersAuthor 5 of 5NeurIPS 2022 · 163 citations
  5. Leveraging unlabeled data to predict out-of-distribution performanceAuthor 4 of 5ICLR 2022 · 160 citations
  6. Exploring the Limits of Large Scale Pre-trainingAuthor 3 of 4ICLR 2022 · 135 citations
  7. Sharpness-aware Minimization for Efficiently Improving GeneralizationAuthor 4 of 4ICLR 2021 · 1,861 citations
  8. Understanding the failure modes of out-of-distribution generalizationAuthor 3 of 3ICLR 2021 · 205 citations
  9. Deep Learning Through the Lens of Example DifficultyAuthor 3 of 3NeurIPS 2021 · 204 citations
  10. When Do Curricula Work?Author 3 of 3ICLR 2021 · 141 citations
  11. Are wider nets better given the same number of parameters?Author 2 of 3ICLR 2021 · 48 citations
  12. Extreme Memorization via Scale of InitializationAuthor 3 of 3ICLR 2021 · 22 citations
  13. Fantastic Generalization Measures and Where to Find ThemAuthor 2 of 5ICLR 2020 · 705 citations
  14. Observational Overfitting in Reinforcement LearningAuthor 5 of 7ICLR 2020 · 148 citations
  15. Towards Learning Convolutions from ScratchAuthor 1 of 1NeurIPS 2020 · 80 citations
  16. The intriguing role of module criticality in the generalization of deep networksAuthor 2 of 3ICLR 2020 · 59 citations
Dusk over the sea between two cliffs drawn in fine vertical lines