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

Language Modeling and Reasoning on ARC-C, ARC-E, HellaSwag, LAMBADA (LO/LS), PIQA, WinoGrande (zero-shot)

78.6Accuracy (ARC-E)

TWLA

Updated 1mo ago

Evaluation Results

Method
TWLA 2026.06	78.6	51.88	70	75.93	69.07	77.33	74.9	71.1
TWLA 2026.06	73.34	51.83	67.8	65.9	55.5	74.29	68.1	65.25
TWLA 2026.06	68.39	45.05	65.29	61.48	57.17	72.91	63.61	62
TWLA 2026.06	66.58	39.85	68.2	73.24	60.82	74.27	67.17	64.3
QuaRot 2026.06	63.5	34.2	55.2	57.5	35.2	68.6	61.3	53.64
TWLA 2026.06	60.23	34.56	59.83	51.87	45.47	71.38	63.3	55.23
TWLA 2026.06	58.88	34.73	60.81	64.45	53.76	71.27	62.12	58
TWLA 2026.06	57.19	35.83	51.82	40.49	39.97	68.22	59.45	50.42
ResQ 2026.06	56.75	32.94	54.8	50.1	38.7	68.66	60.93	51.84
ResQ 2026.06	46.65	28.5	48.05	45.3	32.1	63.32	52.72	45.23
ResQ 2026.06	42.9	26.96	43.85	23.35	21.75	59.64	55.38	39.12
ResQ 2026.06	42.6	34.6	62.4	70	60.9	55.1	69.4	56.43
QuaRot 2026.06	39.56	25.77	39.94	32.08	23.37	60.12	55.09	39.42
QuaRot 2026.06	35.02	22.87	31.34	10.27	8.48	56.75	51.07	30.83
QuaRot 2026.06	31.61	21.42	29.26	4.81	4.97	53.43	48.3	27.69
PB-LLM 2026.06	29.7	22.78	30.3	0.85	0.5	51.03	48.86	26.29
PT2-LLM 2026.06	29.2	26.21	30.22	0.9	0.65	51.23	52.77	27.31
PB-LLM 2026.06	28.85	23.81	28.65	0.15	0.05	52.18	50.12	26.26
PT2-LLM 2026.06	28.71	22.02	30.1	0.75	0.3	50.83	47.99	25.81
PB-LLM 2026.06	28.2	25.26	29.15	0.7	0.35	50.11	51.46	26.46
PB-LLM 2026.06	28.05	23.63	31.9	0.8	0.65	50.6	49.01	29.23
SliM-LLM 2026.06	27.15	25.17	30.45	0.4	0	50.87	51.07	26.44
SliM-LLM 2026.06	27.15	25.68	28.05	0.05	0	48.69	47.99	25.37
PT2-LLM 2026.06	27.1	26.54	28.95	0	0	49.69	51.22	26.19
PB-LLM 2026.06	27.1	25.77	28.8	0.3	0.4	51.09	51.07	26.36
PB-LLM 2026.06	26.95	24.4	29.5	0.55	0.45	50.44	50.75	26.18
PT2-LLM 2026.06	26.95	25.4	29.59	0.35	0.15	50.22	50.66	26.19
SliM-LLM 2026.06	26.8	25.43	28.1	0.15	0.15	49.24	49.8	25.67
GPTQ 2026.06	26.6	25.94	25.74	0	0	52.07	50.2	25.79
GPTQ 2026.06	26.56	25	26.46	0	0	50.38	51.07	25.64
GPTQ 2026.06	26.39	27.82	25.93	0	0	48.53	48.78	25.35
GPTQ 2026.06	26.39	26.88	26.14	0	0	49.46	52.57	25.92
SliM-LLM 2026.06	26.35	26.28	27.1	0	0	48.75	50.75	25.6
SliM-LLM 2026.06	26.35	25.77	27.4	0	0	50.27	49.57	25.62
GPTQ 2026.06	26.26	27.73	26.35	0	0	49.89	50.36	25.8
PB-LLM 2026.06	26.1	26.54	26.95	0	0	48.69	50.51	25.54
SliM-LLM 2026.06	26.1	24.77	29.8	0.85	0.55	50.89	51.22	26.31
GPTQ 2026.06	25.17	26.37	25.47	0	0	50.38	49.64	25.29
GPTQ 2026.06	24.62	25.6	26.13	0	0	49.29	48.3	24.85