Researchers (2025a) analyzed the optimization dynamics of a single-layer Transformer with normalized ReLU self-attention under in-context learning (ICL) mechanisms, finding that smaller eigenvalues of attention weights preserve basic knowledge, while larger eigenvalues capture specialized knowledge.

Authors

• Person: Not available Organization: arXiv
  A Survey on the Theory and Mechanism of Large Language Models

Sources

• A Survey on the Theory and Mechanism of Large Language Models arxiv.org arXiv via serper

Referenced by nodes (2)

• In-Context Learning concept
• Transformer concept