Machine Unlearning on Enron + News (20% forget set) benchmark leaderboard · SOTA2 Research

Machine Unlearning on Enron + News (20% forget set)

40.73P1: Recall on D_fkRMGDR

Evaluation Results

Method	Links
RMGDR2026.04		40.73	38.87	36.1	37.07	65.12	63.08	32.31	35.55	52.55
RM2026.04		30.73	28.33	38.05	37.22	25.85	21.81	29.09	34.67	52.63
IDKGDR2026.04		21.83	20.07	29.63	29.61	66.71	66.31	33.49	34.99	52.51
IDK2026.04		16.71	15.35	32.56	34.05	67.68	66.43	28.95	35.95	52.63
Target f_target2026.04		14.53	14.53	29.94	29.94	60.6	60.6	27.35	36.38	52.59
RLGDR2026.04		12.44	12.03	39.63	37.56	62.93	60.83	31.33	34.07	52.39
RL2026.04		12.32	13	37.32	37.95	65.85	64.64	29.25	36.36	52.62
GAKLR2026.04		12.2	13.13	31.34	32.84	66.1	64.58	27.35	38.25	52.66
NPOKLR2026.04		12.2	13.13	31.34	32.84	64.58	64.58	27.35	36.38	52.97
Task Vector2026.04		2.92	2.62	12.68	12.82	66.1	64.58	26.49	37.05	52.53
WHP2026.04		1.59	0.7	22.68	23.09	66.71	65.28	28.76	36.09	52.43
WHPGDR2026.04		0.98	0.82	12.32	12.98	65.85	66.49	27.49	35.38	52.37
GA2026.04		0.49	0.43	0.49	0.61	54.27	54.68	26.71	36.19	52.59
NPO2026.04		0.49	0.49	0.24	0.06	56.83	58.22	23.78	36.48	52.47
NPOGDR2026.04		0.37	0.52	0.24	0.06	57.56	57.42	24.55	37.49	53.15
GAGDR2026.04		0.24	0.43	0.37	0.47	53.1	53.3	27.57	34.27	52.54
Retrain f_retrain2026.04		0.02	0.02	4.95	4.95	13.14	13.14	26.69	35.04	55.08
DPO2026.04		0	0	30.61	31.16	65.98	65.46	9.97	24.76	54.21
DPOGDR2026.04		0	0	34.27	35.09	69.51	67.86	10.02	24.83	54.39
DPOKLR2026.04		0	0.06	51.1	51.11	65.24	63.81	11.69	22.84	53.56
RMU2026.04		0	0	0	0	18.32	20.3	8.35	9.51	50.45