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

Lifelong Model Editing on ZsRE

91.09Efficacy

RLEDIT

Updated 21d ago

Evaluation Results

Method
RLEDIT 2026.05	91.09	89.62	41.43	-
RLEDIT 2026.05	91.09	89.62	41.43	-
STABLEEDIT 2026.05	90.61	88.38	48.23	-
STABLEEDIT 2026.05	90.61	88.38	48.23	-
ULTRAEDIT 2026.05	90.07	87.36	49.51	-
ULTRAEDIT 2026.05	90.07	87.36	49.51	-
STABLEEDIT 2026.05	87.39	82.93	50.26	-
STABLEEDIT 2026.05	87.39	82.93	50.26	-
ALPHAEDIT 2026.05	86.1	81.15	26.26	-
ALPHAEDIT 2026.05	86.1	81.15	26.26	-
ULTRAEDIT 2026.05	85.3	80.8	47.38	-
ULTRAEDIT 2026.05	85.3	80.8	47.38	-
RLEDIT 2026.05	84.7	82.79	38.26	-
HiEditrand 2026.04	82.65	78.98	31.95	67.89
HiEditfull 2026.04	82.12	78.4	32.4	68.73
STABLEEDIT 2026.05	82.12	78.39	31.52	-
STABLEEDIT 2026.05	82.12	78.39	31.52	-
HiEditfull 2026.04	82.1	79.99	48.42	75.16
ULTRAEDIT 2026.05	82.03	77.08	45.51	-
HiEditrand 2026.04	81.95	79.63	47.97	74.66
RLEdit 2026.04	81.43	79.49	42.73	70.72
STABLEEDIT 2026.05	80.4	76.65	47.59	-
ULTRAEDIT 2026.05	78.03	72.42	27.05	-
ULTRAEDIT 2026.05	78.03	72.42	27.05	-
RLEDIT 2026.05	72.65	68.45	21.79	-
RLEDIT 2026.05	72.65	68.45	21.79	-
RLEDIT 2026.05	71.62	67.6	30.7	-
RLEDIT 2026.05	71.62	67.6	30.7	-
RLEdit 2026.04	69.11	65.87	25.17	57.66
ALPHAEDIT 2026.05	50.23	43.31	12.54	-
ALPHAEDIT 2026.05	50.23	43.31	12.54	-
Pre-edited 2026.05	44.46	43.55	48.08	-
Pre-edited 2026.05	44.46	43.55	48.08	-
Pre-edited 2026.05	36.76	35.83	38.93	-
Pre-edited 2026.05	36.76	35.83	38.93	-
Pre-edited 2026.04	36.58	35.89	38.64	38.26
FT 2026.04	35.38	32.31	30.46	33.4
Pre-edited 2026.05	34.32	33.39	38.06	-
Pre-edited 2026.04	32.75	31.87	39.46	32.36
Pre-edited 2026.05	27.22	26.42	27.33	-
Pre-edited 2026.05	27.22	26.42	27.33	-
DAFNet 2026.04	23.02	22.17	28.32	23.15
DAFNet 2026.04	21.15	20.41	21.15	21.13
MEND 2026.04	20.78	20.29	12.34	21.11
FT 2026.05	18.51	18.11	5.01	-
FT 2026.05	18.51	18.11	5.01	-
ALPHAEDIT 2026.05	16.32	13.96	1.66	-
AlphaEdit 2026.04	16.22	14.58	3.87	9.1
AlphaEdit 2026.04	15.79	15.32	20.21	13.19
FT 2026.05	15.11	13.55	2.61	-
FT 2026.05	15.11	13.55	2.61	-
MEND 2026.05	15	14.41	0.47	-
FT 2026.05	14.02	10.91	3.39	-
FT 2026.05	13.69	12.43	19.87	-
FT 2026.05	13.69	12.43	19.87	-
RECT 2026.04	11.26	11.25	16.19	9.62
MEMIT 2026.04	10.54	10.4	14.66	9.22
PRUNE 2026.04	10.46	10.5	14.66	9.46
MALMEN 2026.04	9.3	8.87	12.9	11.22
FT 2026.04	9.16	8.11	1.95	8.82
MALMEN 2026.04	5.09	4.74	0.72	4.26
MALMEN 2026.05	3.86	3.07	0.26	-
MALMEN 2026.05	3.86	3.07	0.26	-
MEND 2026.05	3.33	3.08	0.2	-
MEND 2026.05	3.33	3.08	0.2	-
ROME 2026.04	3.29	3.24	0.66	2.67
MEND 2026.05	2.52	2.55	0.19	-
MEND 2026.05	2.52	2.55	0.19	-
MALMEN 2026.05	1.15	1.06	0.02	-
MALMEN 2026.05	1.15	1.06	0.02	-
MEMIT 2026.05	0.14	0.14	1.4	-
MEMIT 2026.05	0.14	0.14	1.4	-
MALMEN 2026.05	0.02	0.01	0.02	-
MALMEN 2026.05	0.02	0.01	0.02	-
MEMIT 2026.05	0.02	0.02	0.17	-
MEMIT 2026.04	0	0	0.12	0
PRUNE 2026.04	0	0	0	0
RECT 2026.04	0	0	0	0
MEND 2026.04	0	0	0	0.02
ROME 2026.04	0	0	0	0
MEMIT 2026.05	0	0	0	-
ALPHAEDIT 2026.05	0	0	0	-
MEND 2026.05	0	0	0	-
MEMIT 2026.05	0	0	0	-
MEMIT 2026.05	0	0	0	-
ALPHAEDIT 2026.05	0	0	0	-
MEND 2026.05	0	0	0	-
MEMIT 2026.05	0	0	0	-
MALMEN 2026.05	0	0	0	-