# PATTERN Can Be Solved from Memory: Generator-Aware Baselines for Synthetic Node Classification

arXiv: https://arxiv.org/abs/2610.15020 · HTML: https://mlsanity.com/paper/2610.15020/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 38 of its 52 references that are in the map.
- First posted on arXiv 10 Oct 2026.

## Builds on (most cited first, 30 of 38)

- [Fast Graph Representation Learning with PyTorch Geometric](https://mlsanity.com/paper/1903.02428/) (2019)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Do ImageNet Classifiers Generalize to ImageNet?](https://mlsanity.com/paper/1902.10811/) (2019)
- [Recipe for a General, Powerful, Scalable Graph Transformer](https://mlsanity.com/paper/2205.12454/) (2022)
- [Benchmarking Graph Neural Networks](https://mlsanity.com/paper/2003.00982/) (2020)
- [Rethinking Graph Transformers with Spectral Attention](https://mlsanity.com/paper/2106.03893/) (2021)
- [A critical look at the evaluation of GNNs under heterophily: Are we really making progress?](https://mlsanity.com/paper/2302.11640/) (2023)
- [Community Detection and Stochastic Block Models](https://mlsanity.com/paper/1703.10146/) (2017)
- [Structure-Aware Transformer for Graph Representation Learning](https://mlsanity.com/paper/2202.03036/) (2022)
- [Exphormer: Sparse Transformers for Graphs](https://mlsanity.com/paper/2303.06147/) (2023)
- [Graph Inductive Biases in Transformers without Message Passing](https://mlsanity.com/paper/2305.17589/) (2023)
- [Question and Answer Test-Train Overlap in Open-Domain Question Answering Datasets](https://mlsanity.com/paper/2008.02637/) (2020)
- [Directional Graph Networks](https://mlsanity.com/paper/2010.02863/) (2020)
- [Global Self-Attention as a Replacement for Graph Convolution](https://mlsanity.com/paper/2108.03348/) (2021)
- [Rank-normalization, folding, and localization: An improved widehatR for assessing convergence of MCMC](https://mlsanity.com/paper/1903.08008/) (2019)
- [GRPE: Relative Positional Encoding for Graph Transformer](https://mlsanity.com/paper/2201.12787/) (2022)
- [Do We Train on Test Data? Purging CIFAR of Near-Duplicates](https://mlsanity.com/paper/1902.00423/) (2019)
- [Position: Graph Learning Will Lose Relevance Due To Poor Benchmarks](https://mlsanity.com/paper/2502.14546/) (2025)
- [Learning Probabilistic Symmetrization for Architecture Agnostic Equivariance](https://mlsanity.com/paper/2306.02866/) (2023)
- [Can Classic GNNs Be Strong Baselines for Graph-level Tasks? Simple Architectures Meet Excellence](https://mlsanity.com/paper/2502.09263/) (2025)
- [Understanding Isomorphism Bias in Graph Data Sets](https://mlsanity.com/paper/1910.12091/) (2019)
- [Are More Layers Beneficial to Graph Transformers?](https://mlsanity.com/paper/2303.00579/) (2023)
- [Taxonomy of Benchmarks in Graph Representation Learning](https://mlsanity.com/paper/2206.07729/) (2022)
- [Optimal Inference in Contextual Stochastic Block Models](https://mlsanity.com/paper/2306.07948/) (2023)
- [Joint Graph Rewiring and Feature Denoising via Spectral Resonance](https://mlsanity.com/paper/2408.07191/) (2024)
- [Cooperative Sheaf Neural Networks](https://mlsanity.com/paper/2507.00647/) (2025)
- [Gradformer: Graph Transformer with Exponential Decay](https://mlsanity.com/paper/2404.15729/) (2024)
- [Phase transitions in semisupervised clustering of sparse networks](https://mlsanity.com/paper/1404.7789/) (2014)
- [CKGConv: General Graph Convolution with Continuous Kernels](https://mlsanity.com/paper/2404.13604/) (2024)
- [Fixed Aggregation Features Can Rival GNNs](https://mlsanity.com/paper/2601.19449/) (2026)

## Abstract (arXiv)

PATTERN, a synthetic node-classification benchmark still used to rank graph networks, plants a random subgraph in a random host graph and asks which nodes belong to it. Its generator draws 100 patterns once and plants the same ones in every split, so each test graph carries a pattern that also appears in 100 training graphs. A matcher that reads these patterns off the training labels and looks them up in each test graph reaches 99.2 class-weighted accuracy with no trained parameters, against 89.8 for the best published model. Inference under the generator without the patterns reaches 86.8. Two published models, CKGConv and GRIT, score lower on patterns they have not seen: on new graphs with fresh patterns both fall below this pattern-blind predictor, and withholding patterns from CKGConv's training lowers its accuracy on the test graphs that carry them. Training CKGConv with a new pattern in every graph raises its accuracy on fresh patterns by about 2 points, to the level of the pattern-blind predictor, at some cost on the reused patterns. The same kind of inference under the generator matches the best published result on CLUSTER and, without any training, scores above every published result we found on Minesweeper's released board. We recommend that synthetic benchmarks report scores on fresh templates beside reused ones, publish baselines built from their generators, and state how their metric is weighted and pooled.