Over five matched evaluation rounds, SkillSonar reduces ASR across all individual risk families, achieving reductions of 50.0 pp on Resource & Reliability, 42.9 pp on Execution Safety, 35.3 pp on Specification Integrity, 18.3 pp on Operational Side Effects, 61.5 pp on Privacy & Data Flow (OOD), and 44.8 pp on Capability Control (OOD).
Source: source-paper:Table 10, page 22
Reported and observed measurements
attack_success_rate
m-t10-resource-reliability-noguard-asr
Reported 61.3 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-resource-reliability-skillsonar-asr
Reported 11.3 percentage_points
Observed — percentage_points
attack_success_rate_reduction
m-t10-resource-reliability-reduction
Reported 50 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-execution-safety-noguard-asr
Reported 48.6 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-execution-safety-skillsonar-asr
Reported 5.7 percentage_points
Observed — percentage_points
attack_success_rate_reduction
m-t10-execution-safety-reduction
Reported 42.9 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-specification-integrity-noguard-asr
Reported 41.2 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-specification-integrity-skillsonar-asr
Reported 5.9 percentage_points
Observed — percentage_points
attack_success_rate_reduction
m-t10-specification-integrity-reduction
Reported 35.3 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-operational-side-effects-noguard-asr
Reported 36.7 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-operational-side-effects-skillsonar-asr
Reported 18.3 percentage_points
Observed — percentage_points
attack_success_rate_reduction
m-t10-operational-side-effects-reduction
Reported 18.3 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-privacy-data-flow-noguard-asr
Reported 67.7 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-privacy-data-flow-skillsonar-asr
Reported 6.2 percentage_points
Observed — percentage_points
attack_success_rate_reduction
m-t10-privacy-data-flow-reduction
Reported 61.5 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-capability-control-noguard-asr
Reported 59.4 percentage_points
Observed — percentage_points
attack_success_rate
m-t10-capability-control-skillsonar-asr
Reported 14.5 percentage_points
Observed — percentage_points
attack_success_rate_reduction
m-t10-capability-control-reduction
Reported 44.8 percentage_points
Observed — percentage_points
Assessments (0)
No immutable Assessment has been published for this Claim yet.
Runs (0)
No execution has been linked to this Claim yet.