Scale
- • Very long conversations averaging 300 turns and 9K tokens across as many as 35 sessions.
Memory benchmark
LoCoMo
A dataset and evaluation benchmark for very long-term conversational memory across question answering, event summarization, and multimodal dialogue generation.
Tasks
- • Question answering
- • Event summarization
- • Multimodal dialogue generation
What it measures
- • Long-range recall
- • Temporal reasoning
- • Causal reasoning
- • Multimodal memory
Original research
Evaluating Very Long-Term Conversational Memory of LLM Agents
This is the paper that introduced LoCoMo.
arXiv:2402.17753 ↗Used in the field
Papers evaluated on LoCoMo
BenchmarkMemory Architecture
AdaMem: Adaptive User-Centric Memory for Long-Horizon Dialogue Agents
Shannan Yan, Jingchen Ni et al.
· 2026
AdaMem organizes dialogue history into Working Memory, Episodic Memory, Persona Memory, and Graph Memory coordinated by a Memory Agent, Research Agent, and Working Agent. On LoCoMo with GPT-4.1-mini, AdaMem achieves 44.65 F1 overall, beating the best baseline LangMem at 41.76 F1 by +2.89.
Agent MemoryLong-Term Memory
Adaptive Memory Admission Control for LLM Agents
Guilin Zhang, Wei Jiang et al.
· 2026
A-MAC scores candidate memories using Utility, Confidence, Novelty, Recency, and Type Prior combined by a learned linear admission policy with Algorithm 1 A-MAC Memory Admission. On the LoCoMo benchmark, A-MAC achieves F1 0.583 and 2644 ms latency, improving F1 by 0.042 and reducing latency by 1187 ms compared to A-mem.
Benchmark
AgentKVShift: Efficient KV Cache Reuse for Agentic Memory Systems
Nilesh Prasad Pandey, Jason Kong et al.
arXiv 2026 · 2026
AgentKVShift reuses KV cache per retrieved memory unit using Probe Selection, Mean Estimation, and AgentKVShift Correction to adjust stale keys and values. On LoCoMo with AMem and Qwen2.5-3B-Instruct, AgentKVShift achieves 0.319 F1 at recompute ratio 0.1, compared to 0.178 for CacheBlend and 0.125 for ProphetKV.
Benchmark
AgentMemBench: A Systematic Benchmark for Evaluating Long-Term Memory Management Strategies in Conversational AI Agents
Ahmed Cherif
arXiv 2026 · 2026
AgentMemBench wires five strategies—In-Context Windowing (ICW), External Key-Value Store (EKV), Graph-Based Episodic Memory (GEM), Compression-Based Summarisation (CBS), and Web-Augmented Memory (WAM)—into a shared Store Read Generate interface over multi-session dialogues. AgentMemBench’s main result is that EKV achieves macro Recall@5 0.792 on LoCoMo, MultiDoc2Dial, and MSC, a +0.324 gain over ICW’s 0.468, while also leading MRR, Answer F1, and Faithfulness.
Long-Term Memory
All-Mem: Agentic Lifelong Memory via Dynamic Topology Evolution
Can Lv, Heng Chang et al.
· 2026
All-Mem organizes long-term agent memory through Online/Offline Decoupling, Agentic Topology Consolidation, and Topology-Aware Retrieval over a topology-structured memory bank. On LoCoMo, All-Mem reaches 54.63 4o-J versus Mem0’s 48.91, and on LongMemEval-S All-Mem reaches 60.20 4o-J versus Mem0’s 55.80.
Agent MemoryLong-Term Memory
AMA: Adaptive Memory via Multi-Agent Collaboration
Weiquan Huang, Zixuan Wang et al.
· 2026
AMA orchestrates four agents — the Constructor, Retriever, Judge, and Refresher — to build Raw Text, Fact Knowledge, and Episode Memory and route queries adaptively across these granularities. On the LoCoMo benchmark with GPT-4.1-mini, AMA achieves an overall LLM Score of 0.805 compared to Nemori’s 0.774, while reducing token consumption by approximately 80% relative to FullContext.