Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Evidence Retrieval on Synthetic benchmark B=256 197 episodes, 907 queries
Loading...
8.4
Hit@2
OSU-Mem
-0.336
1.932
4.2
6.468
Jun 19, 2026
Hit@2
Recall
Updated 25d ago
Evaluation Results
Method
Method
Links
Hit@2
Recall
OSU-Mem
Construction principle...
2026.06
8.4
22.1
Flat + MMR
2026.06
5.2
15.8
Coverage-Aware
2026.06
4.9
15.7
Flat Vector
2026.06
4.5
15.6
Disjoint Hierarchy
2026.06
4.4
15
Tree Abstraction
2026.06
4.4
15.7
OSU-Mem (disjoint)
Construction principle...
2026.06
3.4
13.8
Periodic Summary
2026.06
1.3
9.3
Budget-Trunc Replay
2026.06
0
14.9
Sliding Window
2026.06
0
13.6
Feedback
Search any
task
Search any
task