Search papers, labs, and topics across Lattice.
The paper introduces RADAR, a role-anchored multi-agent debate framework, to detect half-truths by reasoning about what is omitted rather than explicitly stated. RADAR uses a Politician and Scientist agent to debate retrieved evidence, moderated by a Judge, with a dual-threshold controller for early termination. Experiments show RADAR outperforms single- and multi-agent baselines in omission detection accuracy and reasoning cost across datasets and backbones.
Uncover misleading half-truths by pitting a Politician agent against a Scientist agent in a debate moderated by a Judge, revealing what's left unsaid.
Half-truths, claims that are factually correct yet misleading due to omitted context, remain a blind spot for fact verification systems focused on explicit falsehoods. Addressing such omission-based manipulation requires reasoning not only about what is said, but also about what is left unsaid. We propose RADAR, a role-anchored multi-agent debate framework for omission-aware fact verification under realistic, noisy retrieval. RADAR assigns complementary roles to a Politician and a Scientist, who reason adversarially over shared retrieved evidence, moderated by a neutral Judge. A dual-threshold early termination controller adaptively decides when sufficient reasoning has been reached to issue a verdict. Experiments show that RADAR consistently outperforms strong single- and multi-agent baselines across datasets and backbones, improving omission detection accuracy while reducing reasoning cost. These results demonstrate that role-anchored, retrieval-grounded debate with adaptive control is an effective and scalable framework for uncovering missing context in fact verification. The code is available at https://github.com/tangyixuan/RADAR.
NUS