Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-agent attribution on Synthetic Additive Benchmark
Loading...
0.855
MAE
Who&When
0.8432
0.92285
1.0025
1.08215
May 12, 2026
MAE
Cosine Similarity
Efficiency Error
Top-k Kendall Tau
Updated 21d ago
Evaluation Results
Method
Method
Links
MAE
Cosine Similarity
Efficiency Error
Top-k Kendall Tau
Who&When
type=LLM-as-Judge, var...
2026.05
0.855
17.7
3.664
4
MAST
type=LLM-as-Judge, var...
2026.05
0.882
8
3.811
8
Ours
method=path integral,...
2026.05
1.15
100
1.45
100
LOO
description=leave-one-out
2026.05
1.15
100
2.33
100
Sampled Shapley
permutation samples=200
2026.05
1.15
100
6.79
100
Banzhaf
coalition samples=200
2026.05
1.15
100
6.89
100
Feedback
Search any
task
Search any
task
