From Token Lists to Graph Motifs: Weisfeiler-Lehman Analysis of Sparse Autoencoder Features

Documenti analoghi

• Generalizing Weisfeiler-Lehman Kernels to Subgraphs
  di: Kim, Dongkwan, et al.
  Pubblicazione: (2024)
• Weisfeiler-Lehman goes Dynamic: An Analysis of the Expressive Power of Graph Neural Networks for Attributed and Dynamic Graphs
  di: Beddar-Wiesing, Silvia, et al.
  Pubblicazione: (2022)
• Attacking Graph Neural Networks with Bit Flips: Weisfeiler and Lehman Go Indifferent
  di: Kummer, Lorenz, et al.
  Pubblicazione: (2023)
• Patch-Effect Graph Kernels for LLM Interpretability
  di: Fernandez-Boullon, Ruben, et al.
  Pubblicazione: (2026)
• How Hard Is It for Message-Passing GNNs to Simulate One Weisfeiler-Lehman Color-Refinement Step?
  di: Cui, Guanyu, et al.
  Pubblicazione: (2024)