Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Leave-one-corpus-out transfer on tau2-bench
Loading...
72.5
AUC
run size (flat)
64.284
66.417
68.55
70.683
Jun 22, 2026
AUC
95% CI (AUC)
P-value (> 0.5)
Updated 1mo ago
Evaluation Results
Method
Method
Links
AUC
95% CI (AUC)
P-value (> 0.5)
run size (flat)
Features=flat
2026.06
72.5
66.5
0.001
flat plus dependency layer (f+d)
Features=f+d
2026.06
64.9
58.5
0.001
size-normalized dependency layer (dep)
Features=dep
2026.06
64.6
58.4
0.001
Feedback
Search any
task
Search any
task