Machine Unlearning on Office-31 D → A (test) benchmark leaderboard · SOTA2 Research

Machine Unlearning on Office-31 D → A (test)

78.6Retained Performance D->AUnlearned(+)

Evaluation Results

Method	Links
Unlearned(+)2026.04		78.6	89	42
Finetune2026.04		77.2	79.4	43
Original (SF(DA)2)2026.04		77.1	84.8	42
Retrain2026.04		76	0	76
UC-SCADA-UL2026.04		72.1	0	72
PADA2026.04		66.3	90.7	35
SHOT2026.04		65.9	69.6	39
Nabla Tau2026.04		65.5	36.9	48
ZSMU2026.04		63.3	92.4	33
Lipschitz2026.04		58.3	37.4	42
UNSIR2026.04		58	1.2	57