Language Modeling on Shakespeare Low Resource benchmark leaderboard · SOTA2 Research

Language Modeling on Shakespeare Low Resource

10.46PPLProposed

Evaluation Results

Method	Links
Proposed2026.01		10.46	—	34.59	70.64	50.6	67.3
Proposed+SE2026.01		10.61	—	34.39	70.44	50.41	59.44
Proposed+alpha2026.01		10.61	—	37.2	72.39	52.66	68.23
CosReg2026.01		10.64	—	31.77	68.49	48.17	54.5
Baseline2026.01		10.65	—	31.47	68.83	48.03	41.73
Adv2026.01		10.81	—	31.37	68.57	47.89	25.14
AGG2026.01		11.82	—	29.15	66.91	46.05	52.98
AGG2026.01		12.24	—	28.75	65.39	45.39	26
Proposed+SE2026.01		13.91	—	45.44	78.27	59.86	71.09
Proposed2026.01		17.3	—	37.14	72.04	52.55	76.51
CWT2026.01		17.7	—	19.49	48.96	34.42	64.17
Proposed2026.01		29.9	—	37.16	72.29	52.69	77.67
Proposed+alpha+SE2026.01		38.5	—	48.85	81.32	62.99	82.98
Proposed+alpha2026.01		50.38	—	37.59	72.44	52.96	78.12
Proposed2026.01		100.66	—	5.94	40.65	21.62	93.52