Poetry Understanding Evaluation on Chinese Poetry Understanding Dataset (test) benchmark leaderboard · SOTA2 Research

Poetry Understanding Evaluation on Chinese Poetry Understanding Dataset (test)

97.25Content ScoreQwen-Turbo-Poller

Evaluation Results

Method	Links
Qwen-Turbo-Poller2026.06		97.25	90.88	93.75	91.38	90.88	89	94.38	93.88	92.66
GPT-42026.06		95.25	92.25	97	93.13	88.75	90.38	96.5	94.38	93.46
Claude3.5-Poller2026.06		91.9	88	93.73	84.95	89.35	91.88	92.68	95.18	90.96
DeepSeek2026.06		91.45	89.93	92.58	89.2	86.85	96.38	92.75	94.35	91.69
Claude3.52026.06		91.45	88.55	91.55	84.15	88.45	89.25	92.83	94.5	90.09
GPT-4-Poller2026.06		90.88	89	93.88	89.38	86.38	83	92.63	91.63	89.6
Qwen-Turbo2026.06		90	88.25	90	90	80	83	90	90	87.66
DeepSeek-Poller2026.06		89	88.93	90.5	86.8	82.9	87.83	90.15	92.65	88.6
Human2026.06		80.33	79.05	81.18	77.83	76.15	79.4	78.8	79.88	79.08