Scale
- • Human-authored, interdependent task sequences evaluated through Memory-Agent-Environment loops.
Memory benchmark
MemoryArena
An evaluation gym for agent memory in interdependent multi-session tasks where agents must learn from earlier actions and reuse that memory later.
Tasks
- • Web navigation
- • Preference-constrained planning
- • Progressive information search
- • Sequential formal reasoning
What it measures
- • Memory-guided action
- • Cross-session learning
- • Task success
- • Experience reuse
Original research
MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks
This is the paper that introduced MemoryArena.
arXiv:2602.16313 ↗Used in the field
Papers evaluated on MemoryArena
Agent Memory
Agent Memory: Characterization and System Implications of Stateful Long-Horizon Workloads
Yasmine Omri, Ziyu Gan et al.
arXiv 2026 · 2026
Agent Memory decomposes agent workloads into ingestion, memory construction, storage, retrieval, prompt assembly, generation, and maintenance, and classifies ten systems across four paradigms. Agent Memory’s profiling on MemoryAgentBench and MemoryArena reveals over 47× spread in lifecycle energy per correct answer and two orders of magnitude differences in serving latency across paradigms.
SurveyRAGAgent Memory
Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers
Pengfei Du
· 2026
Memory for Autonomous LLM Agents decomposes agent memory into a POMDP-grounded write–manage–read loop, a three-dimensional taxonomy, and five mechanism families spanning context compression, retrieval stores, reflection, hierarchical virtual context, and policy-learned management. Memory for Autonomous LLM Agents synthesizes results like Voyager’s 15.3× tech-tree speedup and MemoryArena’s 80%→45% drop to show that memory architecture often matters more than backbone choice.