Scientific Paper Revision on ReviseQA benchmark leaderboard · SOTA2 Research

Scientific Paper Revision on ReviseQA

56.42LC Win Rate (Title)DeepSeek-v3-671B

Evaluation Results

Method	Links
DeepSeek-v3-671B2025.05		56.42	65.71	68.32	74.12	72.11	64.83	67.7
XtraGPT-14B2025.05		55.29	59.43	50.9	59.43	57.87	52.11	55.49
Qwen2.5-72B-Instruct2025.05		50.35	56.78	64.75	67.13	64.59	60.34	61.61
GPT-4o-Mini2025.05		48.8	47.43	55.73	66.07	45.67	39.03	51.75
QwQ-32B2025.05		46.58	85.34	81.99	83.82	82.64	95.69	80.86
Qwen2.5-7B-Instruct2025.05		39.93	45.14	45.64	39.28	33.87	31.17	40.8
DeepSeek-R1-Distill-Llama-8B2025.05		37.98	15.19	10.87	9.64	9.73	9.38	13.94
Qwen-QWQ-32B-Preview2025.05		37.83	34.57	32.13	40.58	30.04	32.91	34.22
Qwen2.5-1.5B-Instruct2025.05		36.07	30.87	25.8	21.34	24.18	24.27	26.8
Llama-3.1-8B-Instruct2025.05		34.78	30.64	35.31	41.6	40.29	18.36	33.51
GPT-3.5-Turbo2025.05		25.73	23.99	21.52	23.16	30.97	17.39	24.24
Llama-3.2-3B-Instruct2025.05		19.93	6.45	9.35	3.8	8.26	4.64	8.73
Phi-42025.05		4.57	44.78	50.9	52.54	26.33	42.01	40.71