Deceptive Grounding: Entity Attribution Failure in Clinical Retrieval-Augmented Generation

A new study reveals that clinical retrieval-augmented generation (RAG) systems can fail by attributing evidence from one drug to another, a phenomenon termed 'deceptive grounding.' Testing 13 models, researchers found deceptive grounding rates up to 86.7% in domain-specific models and 7.8% in a deployed system, with existing evaluation metrics unable to detect the failure. The authors propose entity-attribution verification, which achieves 97.0% precision and 98.7% recall in detecting the issue.

arXiv:2607.09349v1 Announce Type: new Abstract: Retrieval-augmented generation evaluation checks whether model claims are factually grounded in retrieved documents. It does not check whether retrieved evidence is attributed to the correct entity. A clinical RAG response can pass every automated check zero hallucinations, near-perfect faithfulness, real citations while presenting drug Y's clinical evidence as evidence about queried drug X. We term this deceptive grounding DG : a failure invisible to faithfulness, hallucination, and citation checks because every claim is sourced from a real document, about the wrong entity. Using a controlled factorial benchmark across 13 models, we find DG rates spanning 8-87% at peak adversarial conditions. Medical and biomedical fine-tuned models reach up to 86.7%; domain specialization amplifies the failure rather than mitigating it. A controlled ablation identifies the mechanism: removing entity-specific clinical evidence from retrieved documents eliminates entity-attribution failure entirely, shifting all failures to confabulation. The two failure modes respond to the same trigger, taking different paths. Production measurement across 740 drug-disease pairs finds 7.8% overall DG in a deployed RAG system, rising to 13.6% for recently approved drugs. Entity-attribution verification checking that cited evidence applies to the queried entity detects DG at 97.0% precision and 98.7% DG recall IPW-adjusted human gold standard ; no existing framework implements it.