Researchers have developed a new benchmark dataset called EstGraph to evaluate the capabilities of Large Language Models (LLMs) on large-scale graph property estimation. The benchmark addresses the limitation of existing graph datasets, which are too small for LLMs due to context length constraints. EstGraph includes four tasks and utilizes random walk sampling to effectively convey information about graphs with up to millions of nodes to LLMs. AI
Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →
IMPACT Establishes a new evaluation framework for LLMs on complex graph data, potentially improving their applicability in domains with large, interconnected datasets.
RANK_REASON The cluster contains an arXiv preprint detailing a new benchmark dataset for evaluating LLMs on graph property estimation.