AI RESEARCH

Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory

arXiv CS.AI

ArXi:2605.29630v1 Announce Type: cross End-to-end agent-memory benchmarks report a single hit per retriever, confounding lexical leakage (uncontrolled query/gold/distractor entity overlap) with tag-mixing (preferences, services, tools averaged together). We propose entity-collision, a system-agnostic protocol that pins the BM25 floor by construction -- every distractor shares the answer's entity tokens -- and stratifies queries by discriminator tag, so any lift over BM25 is attributable to the embedder. Applied to an open-source agent-memory testbed across 5.