Replies: 4 comments
-
|
Locomo测试的过程中发现一个问题,就是 这给我们带来一个值得注意的点就是,测试代码最好是在相应的数据集的repo里拉取,那么也就是一个数据集一个测试指标,一份代码 例如F1 Score Category 1: Multi-hop(多跳问题) Category 2, 3, 4: Single-hop/Temporal/Open-domain Category 5: Adversarial(对抗性问题) |
Beta Was this translation helpful? Give feedback.
-
|
操作不能太多,也不能太少 记忆前操作-Normalization Strategy
3.复合结构化(这是针对多层记忆结构的) 记忆后操作-Consolidation Policy回忆前操作-Query Formulation Strategy回忆后操作-Context Integration Mechanism |
Beta Was this translation helpful? Give feedback.
-
|
复现进度: |
Beta Was this translation helpful? Give feedback.
-
|
复现prompt 请你首先了解SAGE/packages/sage-benchmark/src/sage/benchmark/benchmark_memory/experiment/memory_test_pipeline.py的测试逻辑 |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
此帖主要记录SAGE-Mem开发过程中遇到的一些问题
Beta Was this translation helpful? Give feedback.
All reactions