CoMemBench: Benchmarking Collaborative Memory Boundaries across Multi-Agent Workflow Topologies
CoMemBench is introduced, an execution-grounded benchmark for collaborative memory sharing and isolation across multi-agent workflow topologies, and reveals a sharing-isolation trade-off: broader context improves information availability but can weaken isolation, while system rankings shift across topologies and artifa...
Sen Zhao, Rui-Qi Kong, Zuyu Zhang et al.
· 1 citation