Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Evidence Retrieval on Synthetic benchmark B=128 197 episodes, 907 queries
Loading...
2.7
Hit@2
OSU-Mem
-0.108
0.621
1.35
2.079
Jun 19, 2026
Hit@2
Recall
Updated 25d ago
Evaluation Results
Method
Method
Links
Hit@2
Recall
OSU-Mem
Construction principle...
2026.06
2.7
13
Flat + MMR
2026.06
2.2
8.5
Flat Vector
2026.06
1.6
8.8
Coverage-Aware
2026.06
1.4
8.6
Disjoint Hierarchy
2026.06
1.4
8.4
Tree Abstraction
2026.06
1.2
9.3
OSU-Mem (disjoint)
Construction principle...
2026.06
0.8
7.5
Budget-Trunc Replay
2026.06
0
0.4
Sliding Window
2026.06
0
0
Periodic Summary
2026.06
0
4.1
Feedback
Search any
task
Search any
task