Share your thoughts, 1 month free Claude Pro on usSee more

Jailbreaking on EHRAgent eICU

57.93Success Rate (SR)

Non-attack

Updated 10d ago

Evaluation Results

Method	Links
Non-attack 2026.04		57.93	-	-	-	98.1
JailAgent 2026.04		57.24	95.69	91.21	90	97.76
BadChain 2026.04		51.72	80.34	78.62	76.55	79.66
AgentPoison 2026.04		51.38	87.76	81.03	80.34	96.38
Non-attack 2026.04		50.86	-	-	-	80.34
JailAgent 2026.04		50.52	91.72	86.55	85.86	79.48
Non-attack 2026.04		49.66	-	-	-	69.31
BadChain 2026.04		48.1	86.21	78.1	77.41	95.52
JailAgent 2026.04		47.93	93.45	91.9	91.38	69.31
AgentPoison 2026.04		47.76	80.86	77.41	75.69	76.9
PAIR 2026.04		47.24	73.97	70.52	67.07	93.79
JailAgent 2026.04		46.55	82.59	78.62	76.72	81.03
Non-attack 2026.04		46.38	-	-	-	63.28
Non-attack 2026.04		46.38	-	-	-	80.17
AgentPoison 2026.04		46.21	87.76	85.52	82.59	62.07
AgentPoison 2026.04		45.69	86.21	79.14	77.41	63.28
BadChain 2026.04		45.34	86.72	83.97	82.07	61.38
JailAgent 2026.04		45.34	91.21	89.31	89.48	62.59
Non-attack 2026.04		44.48	-	-	-	61.72
JailAgent 2026.04		43.79	90.34	84.31	80.86	61.21
BadChain 2026.04		43.28	80.52	75.69	75.52	79.31
AgentPoison 2026.04		42.93	75.86	71.21	70.69	78.79
PAIR 2026.04		42.76	74.66	72.76	70.86	55.17
BadChain 2026.04		42.24	85.86	76.55	74.66	57.93
PAIR 2026.04		42.24	70.86	68.97	65	67.59
AgentPoison 2026.04		42.07	84.14	78.97	78.45	56.21
BadChain 2026.04		40.86	83.62	77.59	77.93	53.28
PAIR 2026.04		39.48	66.72	61.9	58.45	77.93
PAIR 2026.04		36.72	75.34	69.14	69.14	52.76
PAIR 2026.04		34.48	67.93	65	62.41	49.48