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

Content Moderation on Multi-category Chinese Content Moderation Dataset 1.0 (test)

89.7Politics Accuracy

CARO (Llama3-8B)

Updated 5d ago

Evaluation Results

Method	Links
CARO (Llama3-8B) 2026.04		89.7	94.5	97.4	81.4	97.2	72.3	88.8
CARO (Qwen2.5-7B) 2026.04		89.5	93.5	97.8	81.2	98.4	74.6	89.2
Naive SFT 2026.04		82.8	94.4	93.2	70.7	98.6	63.7	84.2
DeepSeek V3 2026.04		79	90.3	89.8	70.5	95	62.5	80.3
CoT SFT+RL 2026.04		78.3	90	91	72.3	95.4	66.3	81.6
CoT SFT 2026.04		77.3	90.6	89.2	65.9	96.7	65.4	80
QwQ-32B 2026.04		75.4	69.6	72	60.7	84.9	54.6	69.1
Class-RAG 2026.04		74.9	83.7	91	68	90	56	75.5
DeepSeek R1 2026.04		72.7	91.4	86.1	64.6	94.3	59.7	77.1
Agentic ICL 2026.04		66.8	87	77.7	67.6	91.8	54.6	72.9
Qwen2.5-32B-Instruct 2026.04		59.1	91.1	84.4	67.9	95.4	54.2	74.3
GPT-4 2026.04		58.6	88.7	79.8	64.3	92.7	56.8	72.3
LLaMA3-8B 2026.04		58.5	55.9	81	65.2	90.6	44.2	67.5
Qwen2.5-7B-Instruct 2026.04		54.9	81.9	70	60.1	84.3	48.8	64.3
LLaMA-Guard-3-8B 2026.04		12	74.1	41.8	45.7	29.4	35.6	39.7