What Improves the Generalization of Graph Transformers? A Theoretical Dive into the Self-attention and Positional Encoding

Documenti analoghi

• Learning on Transformers is Provable Low-Rank and Sparse: A One-layer Analysis
  di: Li, Hongkang, et al.
  Pubblicazione: (2024)
• When is Task Vector Provably Effective for Model Editing? A Generalization Analysis of Nonlinear Transformers
  di: Li, Hongkang, et al.
  Pubblicazione: (2025)
• Improving Transformers using Faithful Positional Encoding
  di: Idé, Tsuyoshi, et al.
  Pubblicazione: (2024)
• Training Nonlinear Transformers for Chain-of-Thought Inference: A Theoretical Generalization Analysis
  di: Li, Hongkang, et al.
  Pubblicazione: (2024)
• Can Mamba Learn In Context with Outliers? A Theoretical Generalization Analysis
  di: Li, Hongkang, et al.
  Pubblicazione: (2025)