#group self-consistency

LifelongAgentBench introduces a novel benchmark for evaluating continuous learning in LLM-based agents, focusing on knowledge retention and adaptation across sequential tasks in dynamic environments.