When Users Don't Ask: Benchmarking Context-Driven Memory Retrieval in Conversational Agents

We further find that strong retrieval does not fully trans- late into response quality, and that implicit queries exhibit silent grounding, where mem- ory improves contextual grounding without ex- plicitly surfacing the gold fact. It comes from a paper posted to arXiv on 4 September 2026. Large language models (LLMs) are increas- ingly deployed as long-horizon conversational agents, motivating growing interest in mem- ory systems. However, existing benchmarks primarily evaluate memory through QA-style probing rather than in-situ conversational usage. We introduce LOCOMO-CONV, a conversa- tional memory benchmark derived from Lo- CoMo with four query styles: dialog, implicit, counterfactual, and composed. Across five rep- resentative memory systems, we evaluate both retrieval recall and end-to-end response qual- ity. Our experiments show that conversational framing exposes substantial retrieval gaps over- looked by QA benchmarks, especially on im- plicit and composed queries, which multi-facet query rewriting narrows for raw-turn mem- ory but not abstractive memory. These results point to reasoning-based memory elaboration as a promising direction, and we release aux- iliary supportive_memory annotations captur- ing conversationally useful context beyond the original gold evidence.

Why it counts

We further find that strong retrieval does not fully trans- late into response quality, and that implicit queries exhibit silent grounding, where mem- ory improves contextual grounding without ex- plicitly surfacing the gold fact.

Sources

Primary source: primary source

What is not known

This brief does not claim independent replication. Claims that appear only on X and not in the primary source stay unknown.