Legal Reasoning Graph Benchmark

GRBench is a benchmark dataset for legal reasoning tasks, structured as a graph. The dataset was created by yuyangbai and last updated in February 2026. Specific details on row count, column count, and data size are unavailable.

Use Cases

• Benchmark graph neural networks on legal document relationships using the graph structure.
• Evaluate legal reasoning models on tasks defined by the 'Legal Reasoning' tags.
• Analyze the connectivity and properties of the legal graph for knowledge representation research.

Strengths

• Dataset is explicitly tagged for 'Legal Reasoning' and 'Benchmarking', indicating a clear purpose.
• Last update recorded as February 2026, suggesting recent maintenance.

Limitations

• Critical metadata such as row count, column definitions, and sample data are unavailable, hindering initial assessment.
• The description references a multi-file structure requiring concatenation, indicating potential complexity in data assembly.

Provenance

Source

Copy of GRBench from PeterJinGo on Hugging Face.

Collection Method

null

Time Range

null

Freshness

Last updated 2026-02-17.

Geography

null