Scenario
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Scenario S3
4.11Mean Score
3
Scenario S2
3.44Mean Violation
3
Scenario S1
4.9Mean Score
3
Scenario S4
41Inference Time (s)
3
Scenario S3
38Inference Time (s)
3
Scenario S2
36Inference Time (s)
3
Scenario S1
33Inference Time (s)
3
Scenario D High Duration 1.0 (test)
75Compliance Score
3
Scenario C High Criticality 1.0 (test case)
0.98Compliance Score
3
Scenario A
30Success Rate (Succ.)
2
Scenario 3
0.342MAE (τ=0.05)
2
Scenario 6 Track design
90.8Safety Metric
2
Scenario 3
54.7Travel Time (s)
2
Scenario 2
77.3Travel Time (s)
2
Scenario 1
44.9Travel Time (s)
2
Scenario 3
68.89Path Length
2
Scenario 2
100.48Path Length
2
Scenario 1
57.24Path Length
2
Scenario S4 C&C communication
1Attack Slowdown
2
Scenario S3 data exfiltration
373Attack Slowdown
2
Scenario S2 network reconnaissance
1Attack Slowdown
2
Scenario S1 SSH brute-force
1Attack Slowdown
2
Scenario C (Zero-Knowledge)
15Success Metric
1
Scenario C Online A/B Test
2.177Revenue
1
Scenario B Online A B Test
1.366Revenue Lift (%)
1