Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
ALFWorld task success on ALFWorld Eval-seen (50 tasks)
Loading...
49
Success Count (Seen Tasks)
TraceRetain-CEM
38.6
41.3
44
46.7
Jun 28, 2026
Success Count (Seen Tasks)
Success Rate 95% CI Lower Bound (Seen Tasks)
Updated 26d ago
Evaluation Results
Method
Method
Links
Success Count (Seen Tasks)
Success Rate 95% CI Lower Bound (Seen Tasks)
TraceRetain-CEM
Capacity=50
2026.06
49
90,100
FIFO
Capacity=50
2026.06
48
8,799
TraceRetain-Linear
Capacity=50
2026.06
48
8,799
Unbounded
2026.06
47
8,498
No memory
2026.06
39
6,587
Feedback
Search any
task
Search any
task