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

Classification on MetaShift

95.17Average Worst Group Accuracy

DFR

Updated 1mo ago

Evaluation Results

Method	Links
DFR 2026.06		95.17	-	-	-	-	75.23
CBDebug 2025.09		89.3	-	-	-	-	93.4
LC 2026.06		88.71	-	-	-	-	76.16
GB ERM 2026.06		88.48	-	-	-	-	75.69
GroupDRO 2026.06		87.62	-	-	-	-	74.07
Reweight Only 2025.09		87.3	-	-	-	-	93.1
CB ERM 2026.06		87.27	-	-	-	-	73.15
Augment Only 2025.09		86.3	-	-	-	-	92.6
TCSL-CS (ERM) 2026.06		84.95	-	-	-	-	79.4
Original 2025.09		84.5	-	-	-	-	92.9
Retrain 2025.09		84.4	-	-	-	-	93.1
TCSL-CS 2026.06		82.64	-	-	-	-	87.27
RGbal 2026.06		78.7	-	-	-	-	89.35
SupER 2025.08		76.8	79.8	78.4	77.6	71.4	-
EL2N (Top) 2026.06		75.46	-	-	-	-	88.54
D2 (EL2N+ResNet) 2026.06		75.46	-	-	-	-	88.25
Remove 2025.09		73.9	-	-	-	-	89
EL2N (Hist) 2026.06		71.76	-	-	-	-	85.19
D2 (SelfSup+CLIP) 2026.06		71.76	-	-	-	-	86.05
SelfSup (Top) 2026.06		71.3	-	-	-	-	85.76
Random 2026.06		71.3	-	-	-	-	85.94
SelfSup (Hist) 2026.06		71.06	-	-	-	-	85.94
D2 (SelfSup+ResNet) 2026.06		70.6	-	-	-	-	85.36
D2 (EL2N+CLIP) 2026.06		70.6	-	-	-	-	85.88
LfF 2025.08		70	77.2	73.9	69.5	59.5	-
CVaR DRO 2025.08		67.5	77.8	72.5	65.1	54.7	-
JTT 2025.08		67.5	76.7	73.2	67.1	53	-
ERM 2025.08		67.3	78.8	75.8	61.9	52.6	-
CnC 2025.08		66.9	81.1	71.4	65.4	49.6	-
GALS 2025.08		66	74.8	68.8	70.6	50	-
SelfSup (Bot) 2026.06		61.34	-	-	-	-	84.84
EL2N (Bot) 2026.06		60.42	-	-	-	-	80.21
CBDebug 2025.09		57.3	-	-	-	-	82.3
Reweight Only 2025.09		56.1	-	-	-	-	81.8
Augment Only 2025.09		55.6	-	-	-	-	82.2
Remove 2025.09		55	-	-	-	-	81.4
Retrain 2025.09		53.3	-	-	-	-	81.2
Original 2025.09		52.4	-	-	-	-	80.9
ProtoPDebug 2025.09		52.4	-	-	-	-	80.9
ERM 2025.08		-	10.3	14.2	411.8	722.1	-
CVaR DRO 2025.08		-	21.3	97.5	237.7	599.7	-
LfF 2025.08		-	15.3	30.3	73.3	258.8	-
GALS 2025.08		-	82.1	197.7	157.7	955.6	-
JTT 2025.08		-	31.2	24.9	128.6	699.8	-
CnC 2025.08		-	2.3	71.5	262.5	769.6	-
SupER 2025.08		-	9.3	13.1	7.7	49.4	-