3 papers
cs.LG2026
PolyGraph Discrepancy: a classifier-based metric for graph generation
Markus Krimmel, Philip Hartout, Karsten Borgwardt +1
Existing methods for evaluating graph generative models primarily rely on Maximum Mean Discrepancy (MMD) metrics based on graph descriptors. While these metrics can rank generative…
cs.LG2026
Fast Graph Generation via Autoregressive Noisy Filtration Modeling
Markus Krimmel, Jenna Wiens, Karsten Borgwardt +1
Existing graph generative models often face a critical trade-off between sample quality and generation speed. We introduce Autoregressive Noisy Filtration Modeling (ANFM), a flexib…
cs.LG2025
Flatten Graphs as Sequences: Transformers are Scalable Graph Generators
Dexiong Chen, Markus Krimmel, Karsten Borgwardt
We introduce AutoGraph, a scalable autoregressive model for attributed graph generation using decoder-only transformers. By flattening graphs into random sequences of tokens throug…