ShardMemo: Masked MoE Routing for Sharded Agentic LLM Memory
This addresses memory scalability and efficiency issues for agentic LLM systems, offering incremental improvements over existing methods.
The paper tackles the bottleneck of external memory in agentic LLM systems by introducing ShardMemo, a budgeted tiered memory service with masked MoE routing, which improves F1 scores by up to +6.82 over baselines and reduces retrieval work by 20.5%.
Agentic large language model (LLM) systems rely on external memory for long-horizon state and concurrent multi-agent execution, but centralized indexes and heuristic partitions become bottlenecks as memory volume and parallel access grow. We present ShardMemo, a budgeted tiered memory service with Tier A per-agent working state, Tier B sharded evidence with shard-local approximate nearest neighbor (ANN) indexes, and Tier C, a versioned skill library. Tier B enforces scope-before-routing: structured eligibility constraints mask ineligible shards before routing or ANN search. We cast shard probing as masked mixture-of-experts (MoE) routing over eligible shards, probing up to $B_{\mathrm{probe}}$ shards via Top-$B_{\mathrm{probe}}$ or adaptive Top-$P$, and use cost-aware gating over profile/observation/session shard families; the router is trained from evidence-to-shard supervision. On LoCoMo, ShardMemo improves over the strongest baseline (GAM) by +5.11 to +6.82 F1 across question categories. Under a fixed-budget routing setting ($B_{\mathrm{probe}}=3$), ShardMemo improves over cosine-to-prototype shard routing by +6.87 F1 while reducing retrieval work (VecScan 521->414, -20.5%) and p95 latency (95->76 ms). On long-context HotpotQA, ShardMemo achieves 63.41/61.88/57.95 F1 at 56K/224K/448K tokens. On ToolBench, Tier C reaches 0.97 Precision@3 and 1.94 StepRed (+10.2% and +7.2% over embedding-similarity retrieval).