This dataset contains the graph instances used in the paper "Unified Multi-Dimensional Benchmark for Complex Graph Reasoning in Large Language Models". It is the data component of the benchmark, intended for use alongside the evaluation code.
edgelists.zip contains one .edge_list JSON file per benchmark task. Each file holds four graph instances of increasing size, keyed by node count.
{
"10": {… See the full description on the dataset page:
https://huggingface.co/datasets/aliwister/graphgym.
