Multilingual Factual Reasoning on mTruthfulQA benchmark leaderboard · SOTA2 Research

Multilingual Factual Reasoning on mTruthfulQA

54.46Score (ar)LLaMA3.1-8B-AdaMCOT

Evaluation Results

Method	Links
LLaMA3.1-8B-AdaMCOT2025.01		54.46	49.68	56.11	50.7	57.49	59.24	57.41	48.45	58.83	49.25	52.78	57.94	68.04	36.35	59.64	57.09	49.12	45.24	50.39	51.81	57.83	71.45	57.25	56.6	60.03	56.61	56.2	45.63	45.82	58.05	61.53	59.14
Qwen2.5-7B2025.01		52.78	42.25	51.87	50.32	51.27	60.59	54.88	20.8	55.53	26.81	44.11	45.29	46.04	18.44	59.13	56.07	29.06	25.07	32.2	30.62	57.2	61.17	52.5	56.98	46.02	46.08	56.33	23.01	22.55	52.6	54.9	59.77
Qwen2.5-7B-AdaMCOT2025.01		47.35	44.81	52.64	51.34	55.33	62.55	56.4	37.21	58.83	31.33	50.06	54.06	69.95	31.65	52.44	56.32	36.43	30.26	37.96	42.38	55.8	76.93	55.2	60.15	56.04	54.17	54.13	31.22	32.77	58.7	47.52	58.5
LLaMA3.1-8B2025.01		46.96	36.11	51.09	48.78	51.9	57.16	54.37	32.56	53.62	48.7	43.73	49.94	47.86	32.01	46.92	54.15	48.38	44.81	47.77	43.41	51.21	57.87	53.53	44.16	48.97	48.78	53.1	49.39	42.84	48.57	52.48	50.13
Qwen2.5-7B-QAlign2025.01		45.8	27.14	46.07	47.25	52.66	59.49	52.09	15.76	52.86	17.24	33.59	41.88	35.15	14.31	52.31	51.85	16.96	15.71	22.38	21.73	52.74	54.06	47.37	50.13	40.49	43.57	48.32	16.55	18.87	44.29	47.39	55.39
Qwen2.5-7B-AutoCAP2025.01		44.18	33.94	45.32	43.75	44.83	53.46	47.02	16.87	50.34	22.15	35.09	42.15	36.72	19.95	50.13	50.86	23.57	21.18	27.25	25.06	50.34	52.61	46.92	50.81	41.67	43.25	48.71	25.9	21.43	45.93	46.66	51.72
LLaMA3.1-8B-QAlign2025.01		43.47	40.46	40.67	36.62	43.4	38.31	38.53	36.05	42.44	42.68	42.3	38.84	41.63	37.01	42.16	39.34	41.45	46.54	45.81	36.82	41.53	40.61	41.21	40.86	38.05	37.74	42.76	42.66	41.13	38.96	38.85	37.39
LLaMA3.1-8B-AutoCAP2025.01		41.01	42.89	50.97	50.66	53.05	49.2	50.44	27.26	52.73	43.64	38.42	48.63	43.71	54.79	50.51	49.55	58.85	43.66	30.63	37.73	51.34	52.54	52.76	45.3	48.2	48.28	52.84	48.32	44.54	45.06	51.97	44.04