LongMemEval-S
A long-term conversational memory evaluation benchmark used to test the performance of agent systems on long-term memory tasks.
Timeline (last 90 days)
agentmemory published a 95.2% R@5 score on the LongMemEval-S benchmark, and it can be reproduced from the test harness in the repository.
MemCompile also shows strong performance on the LongMemEval-S benchmark, indicating that the benefits of query-conditioned compilation generalize to different long-term conversational memory tasks.
Long-term memory evaluation benchmarks (such as LongMemEval and the MemGPT benchmark) assess five dimensions: single-hop information extraction, cross-session reasoning, knowledge updating, temporal reasoning, and abstention judgment.
All Facts (6)
MemGAS在四个长期记忆基准上进行评估：Locomo、LongMemEval-S、LongMemEval-M和LongMTBench Plus
95%UnverifiedMemGAS在LongMemEval-S上构建记忆处理约52.9M input tokens，Summary和Keyword带来的额外内存约27MB，约为原始记忆的10%
85%UnverifiedMemGAS在LongMemEval-S数据集上F1达到20.38，最强基线HIPPO RAG为14.73
55%Unverifiedagentmemory published a 95.2% R@5 score on the LongMemEval-S benchmark, and it can be reproduced from the test harness in the repository.
50%UnverifiedMemCompile also shows strong performance on the LongMemEval-S benchmark, indicating that the benefits of query-conditioned compilation generalize to different long-term conversational memory tasks.
50%UnverifiedLong-term memory evaluation benchmarks (such as LongMemEval and the MemGPT benchmark) assess five dimensions: single-hop information extraction, cross-session reasoning, knowledge updating, temporal reasoning, and abstention judgment.
50%