CWQ

The 'CWQ' dataset is a benchmark that contains complex questions designed to evaluate the reasoning capabilities of models when integrating knowledge from Knowledge Graphs.

Papers using CWQ (2)

Code-on-Graph: Iterative Programmatic Reasoning via Large Language Models on Knowledge Graphs2026Call Me When Necessary: LLMs can Efficiently and Faithfully Reason over Structured Environments2024 · 2 cites