Long-context language modeling on ZeroSCROLLS (test) benchmark leaderboard · SOTA2 Research

Long-context language modeling on ZeroSCROLLS (test)

35.8GovReport ScoreGDWM

Evaluation Results

Method	Links
GDWM2026.01		35.8	9.2	36.5	39.8	87.5	32.8	40.3
GDWM2026.01		35	8.8	35.8	38.6	86.8	32	39.5
GDWM2026.01		34.2	8.4	35.2	37.5	82.2	31.2	38.1
qTTT2026.01		33.2	8.9	33.8	38.4	87.2	30.8	38.7
qTTT2026.01		31.9	8.6	32.1	35.3	84.8	27.5	36.7
GDWM2026.01		31.5	9	28.8	44.2	94.8	51	43.2
GDWM2026.01		30.8	8.6	28.1	43.1	94.3	50.2	42.5
GDWM2026.01		30.2	8.2	27.5	42	93.8	49.5	41.9
GDWM2026.01		29.8	10.2	35.5	13	77.5	28.8	32.5
qTTT2026.01		29.8	9	27	42.4	94.9	49.6	42.2
qTTT2026.01		29.2	8.3	29.5	32.3	81.3	23.7	34
GDWM2026.01		29.1	9.9	34.6	12.6	77	27.9	31.8
GDWM2026.01		28.5	9.6	33.8	12.2	76.5	27	31.3
qTTT2026.01		27.9	8.7	24.5	40.6	93.1	46.2	40.2
qTTT2026.01		26.8	9.7	31.5	12.4	76.5	26.4	30.6
qTTT2026.01		25.3	8.5	22.8	38.5	91.2	42	38.1
qTTT2026.01		25.1	9.3	29.5	11.2	74.1	23.1	28.7
In-context2026.01		24.8	11.2	23.5	10.9	40.8	17	21.4
qTTT2026.01		23.9	8.3	27.3	9.4	72.1	19.9	26.8
In-context2026.01		22.5	8.8	20.1	35.4	90.5	22.9	33.4
In-context2026.01		22.2	6.4	25.8	14.9	48.1	11.8	21.5
Thinking2026.01		21.5	7.6	22.3	9.2	61.8	22.4	24.1
Thinking2026.01		20.8	7.5	25.2	29.8	76.1	8.3	28
Thinking2026.01		18.2	9.8	21.5	19.6	71.8	43.8	30.8