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

Toxicity Mitigation on Toxicity prompts

0.12CLS Toxicity (%)

A-LQR

Updated 3mo ago

Evaluation Results

Method	Links
A-LQR 2026.04		0.12	0.65	8.35
A-LQR 2026.04		0.12	0.7	9.01
ActAdd 2026.04		0.16	0.69	10.81
A-LQR 2026.04		0.18	0.68	12.26
A-LQR 2026.04		0.22	0.68	8.83
S-PID 2026.04		0.28	0.71	8.86
Mean-AcT 2026.04		0.5	0.68	8.7
ActAdd 2026.04		0.56	0.63	12.56
ODESteer 2026.04		0.58	0.64	11.62
S-PID 2026.04		0.6	0.71	8.17
ITI 2026.04		0.64	0.67	9.39
S-PID 2026.04		0.66	0.66	8.1
ODESteer 2026.04		0.68	0.65	10.82
PID-AcT 2026.04		0.7	0.65	7.59
ODESteer 2026.04		0.7	0.66	8.72
Mean-AcT 2026.04		0.72	0.63	7.92
Linear-AcT 2026.04		0.78	0.68	8.94
S-PID 2026.04		0.8	0.7	11.68
PID-AcT 2026.04		0.86	0.69	8.2
Linear-AcT 2026.04		0.92	0.69	8.77
ITI 2026.04		0.96	0.67	10.24
ITI 2026.04		0.96	0.76	8.79
Linear-AcT 2026.04		1.04	0.68	7.46
ActAdd 2026.04		1.1	0.64	11.42
ITI 2026.04		1.62	0.73	-
Linear-AcT 2026.04		3.14	0.7	8.41
Original 2026.04		3.26	0.7	8.3
PID-AcT 2026.04		3.52	0.69	8.99
Mean-AcT 2026.04		3.64	0.67	10.03
Original 2026.04		3.72	0.7	8.61
Original 2026.04		4.16	0.67	8.95
Original 2026.04		5.14	0.67	6.74