Understanding and Minimising Outlier Features in Transformer Training Bobby He
–Neural Information Processing Systems
Despite their widespread use, our understanding of deep neural networks (NNs) and their training dynamics is very much incomplete.
Neural Information Processing Systems
Oct-10-2025, 10:39:23 GMT
- Country:
- Asia > Middle East
- Jordan (0.04)
- Europe
- Poland > Pomerania Province (0.04)
- Switzerland > Zürich
- Zürich (0.04)
- North America
- Dominican Republic (0.04)
- Mexico > Gulf of Mexico (0.14)
- Asia > Middle East
- Genre:
- Research Report
- Experimental Study (1.00)
- New Finding (1.00)
- Research Report
- Industry:
- Education > Educational Setting (0.45)
- Energy (0.46)
- Technology: