Publications

2026

  1. ICML 2026 Oral, WSS@ICML26
    beyond-structural-symmetries-lmc.png
    Beyond Structural Symmetries: Linear Mode Connectivity via Neuron Identifiability
    Vincent Bürgin*, Daniel Herbst*, Ya-Wei Eileen Lin, and Stefanie Jegelka
    International Conference on Machine Learning, Apr 2026
    PDF Code

2025

  1. GFM @ICML 2026
    lost-in-serialization.png
    Lost in Serialization: Invariance and Generalization of LLM Graph Reasoners
    Daniel Herbst*, Lea Karbevska*, Divyanshu Kumar*, Akanksha Ahuja*, Fatemeh Gholamzadeh*, and Fabrizio Frasca
    Workshop on Graph Foundation Models: A New Era for Graph Machine Learning at ICML 2026, Nov 2025
    PDF
  2. ICLR 2025
    Spotlight
    higher-order-gnn-approx-cut-dist.png
    Higher-Order Graphon Neural Networks: Approximation and Cut Distance
    Daniel Herbst, and Stefanie Jegelka
    International Conference on Learning Representations, Feb 2025
    PDF

2024

  1. NeurIPS 2024
    spatio-spectral-gnn.png
    Spatio-Spectral Graph Neural Networks
    Simon Geisler*, Arthur Kosmala*, Daniel Herbst, and Stephan Günnemann
    Advances in Neural Information Processing Systems, Sep 2024
    PDF Code