Wu et al. (2025c) propose a graph theory framework to analyze position bias in multi-layer Transformers, revealing that causal masking inherently biases attention towards earlier positions because tokens in deep layers continuously aggregate context information from earlier tokens.

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)

• Transformers concept
• attention concept