What Breaks Knowledge Graph based RAG? Empirical Insights into Reasoning under Incomplete Knowledge
By: Dongzhuoran Zhou , Yuqicheng Zhu , Xiaxia Wang and more
Potential Business Impact:
Helps computers reason better when information is missing.
Knowledge Graph-based Retrieval-Augmented Generation (KG-RAG) is an increasingly explored approach for combining the reasoning capabilities of large language models with the structured evidence of knowledge graphs. However, current evaluation practices fall short: existing benchmarks often include questions that can be directly answered using existing triples in KG, making it unclear whether models perform reasoning or simply retrieve answers directly. Moreover, inconsistent evaluation metrics and lenient answer matching criteria further obscure meaningful comparisons. In this work, we introduce a general method for constructing benchmarks, together with an evaluation protocol, to systematically assess KG-RAG methods under knowledge incompleteness. Our empirical results show that current KG-RAG methods have limited reasoning ability under missing knowledge, often rely on internal memorization, and exhibit varying degrees of generalization depending on their design.
Similar Papers
Evaluating Knowledge Graph Based Retrieval Augmented Generation Methods under Knowledge Incompleteness
Artificial Intelligence
Makes AI smarter even with missing information.
Human Cognition Inspired RAG with Knowledge Graph for Complex Problem Solving
Machine Learning (CS)
Helps computers solve hard problems by thinking step-by-step.
Towards Self-cognitive Exploration: Metacognitive Knowledge Graph Retrieval Augmented Generation
Information Retrieval
Helps AI learn better by checking its own thinking.