Long-context Language Understanding on LongBench-e (test) benchmark leaderboard · SOTA2 Research

Long-context Language Understanding on LongBench-e (test)

68.42LCC (Language Comprehension Score)HATA

Evaluation Results

Method	Links
HATA2025.06		68.42	10.61	15.65	85.83	54.92	39.61	69.34	22.39	13.44	31.9	0.34	25.06	12.31	34.6
Dense2025.06		67.53	11.89	15.3	85.03	55.03	39.32	69	22.44	13.13	32.01	1.17	24.51	11.76	34.47
HATA2025.06		67.25	99.67	60.19	91.94	51.72	42.35	71.66	55.17	43.82	35.02	12.44	26.07	43.95	53.94
Dense2025.06		67.24	99.67	60.21	91.64	52.36	42.55	71.66	54.82	44.08	35.03	13.19	26.19	44.68	54.1
SnapKV2025.06		66.49	99.67	59.93	91.98	48.6	40.58	60.33	52.82	43.63	26.29	13.04	23.2	36.42	51
MagicPIG2025.06		66.43	10.01	14.69	86.17	55.81	38.94	69	21.7	13.29	31.29	1.08	23.74	11.06	34.09
Quest2025.06		65.14	15.53	13.64	85.18	52.57	39.24	67.57	19.33	12.51	24.83	1.2	16.61	10.93	32.64
H2O2025.06		64.99	94	57.89	92.06	46.7	41.23	65	45.12	40.91	29.4	12.82	24.64	33.75	49.89
StreamingLLM2025.06		64.9	94.33	48.52	79.24	45.58	40.1	51.67	34.13	37.81	22.73	13.15	21.51	24.91	44.51
Loki2025.06		61.29	99.67	59.48	91.45	48.47	41.99	72.33	54.47	44.33	34.74	12.74	25.85	45.15	53.23
Quest2025.06		58.81	99.67	60.03	90.79	46.72	39.75	71.33	51.5	43.9	33.64	13.16	25.69	43.52	52.19
Loki2025.06		58.68	11.97	14.91	85.3	44.41	38.95	69	22.11	13.09	30.51	0.52	23.82	12.82	32.78
MagicPIG2025.06		53.39	98.83	56.28	77.9	42.35	34.28	63.67	49.1	37.84	32.58	9.96	24.57	38.2	47.61
SnapKV2025.06		52.74	9.44	11.78	66.9	45.13	37.56	37	17.44	12.65	13.54	0.08	12.98	7.24	24.96
StreamingLLM2025.06		46.91	4.84	8.22	60.67	35.43	19.5	28	15.51	7.54	19.39	0	18.09	5.43	20.73
H2O2025.06		27.42	2.41	4.19	20.07	16.33	6.74	24.33	3.06	2.19	9.61	0	7.83	0.23	9.57