Aggregate NLP Tasks on Average (Emotion, Irony, Stance, MRPC, RTE) benchmark leaderboard · SOTA2 Research

Aggregate NLP Tasks on Average (Emotion, Irony, Stance, MRPC, RTE)

2.93Delta 1Qwen2.5-14B

Evaluation Results

Method	Links
Qwen2.5-14B2025.11		2.93	0.7
Llama3-8B2025.11		4.38	1.29
Yi1.5-9B2025.11		5.74	0.59
Qwen2.5-7B2025.11		5.83	1.19
Qwen2.5-14B2025.11		6.06	1.72
Ministral-8B2025.11		7.16	0.39
Mistral-12B2025.11		7.19	1.59
Qwen2.5-7B2025.11		7.66	1.95
Yi1.5-6B2025.11		8.88	1.19
Llama3-8B2025.11		10.62	2.09
Yi1.5-9B2025.11		10.71	2.33
Qwen2.5-7B2025.11		12.01	4.74
Mistral-12B2025.11		12.61	2.36
Ministral-8B2025.11		12.84	1.81
Yi1.5-6B2025.11		13.6	1.49
Llama2-13B2025.11		16.27	1.01
Qwen2.5-14B2025.11		17.45	7.19
Llama3-8B2025.11		18.95	6.84
Llama2-13B2025.11		20.92	2.38
Yi1.5-6B2025.11		23.51	7.69
Ministral-8B2025.11		25.17	7.7
Yi1.5-9B2025.11		25.22	10.65
Mistral-12B2025.11		26.5	8.86
Llama2-13B2025.11		31.16	7.07