Sequence Modeling on Long Range Arena (val) benchmark leaderboard · SOTA2 Research

Sequence Modeling on Long Range Arena (val)

62.75ListOps AccuracyLiquid-S4

Evaluation Results

Method	Links
Liquid-S42022.09		62.75	89.02	91.2	89.5	94.8	96.66	87.32
S52022.09		61	86.51	88.26	86.14	87.57	85.25	82.46
S4D-Lin2022.09		60.52	86.97	90.96	87.93	93.96	—	78.39
S4D-LegS2022.09		60.47	86.18	89.46	88.19	93.06	91.95	84.89
S4-LegS/FouT2022.09		60.45	86.78	90.3	89	94.44	—	78.5
S4D-Inv2022.09		60.18	87.34	91.09	87.83	93.78	92.8	85.5
S4-LegS2022.09		59.6	86.82	90.9	88.65	94.2	96.35	86.09
S4 (original)2022.09		58.35	76.02	87.09	87.26	86.05	88.1	80.48
S4-FouT2022.09		57.88	86.34	89.66	89.07	94.46	—	77.9
DSS2022.09		57.6	76.6	87.6	85.8	84.1	85	79.45
H-Transformer-1D2022.09		49.53	78.69	63.99	46.05	68.78	—	61.41
CDIL2022.09		44.05	86.78	85.36	66.91	91.7	—	74.96
Reformer2022.09		37.27	56.1	53.4	38.07	68.5	—	50.56
Luna-2562022.09		37.25	64.57	79.29	47.38	77.72	—	59.37
Nyströmformer2022.09		37.15	65.52	79.56	41.58	70.94	—	57.46
Transformer2022.09		36.37	64.27	57.46	42.44	71.4	—	54.39
BigBird2022.09		36.05	64.02	59.29	40.83	74.87	—	55.01
Longformer2022.09		35.63	62.85	56.89	42.22	69.71	—	53.46
FNet2022.09		35.33	65.11	59.61	38.67	77.8	—	54.42
Sinkhorn Trans.2022.09		33.67	61.2	53.83	41.23	67.45	—	51.23
Performer2022.09		18.01	65.4	53.82	42.77	77.05	—	51.18
Sparse Transformer2022.09		17.07	63.58	59.59	44.24	71.71	—	51.24
Linformer2022.09		16.13	65.9	53.09	42.34	75.3	—	50.55
Linear Trans.2022.09		16.13	65.9	53.09	42.34	75.3	—	50.46
Local Att.2022.09		15.82	52.98	53.39	41.46	66.63	—	46.06
Random2022.09		10	50	50	10	50	50	36.67