小模型通过调用外部工具实现高效内容安全审核
Tool-MCoT: Tool Augmented Multimodal Chain-of-Thought for Content Safety Moderation
- 用大模型生成带工具调用的思维链数据，微调小模型
- 小模型在保持高准确率的同时，推理效率显著提升
- 能智能选择何时调用工具，兼顾速度与精度
在线平台内容增长迅速，亟需能处理多模态复杂输入的内容安全审核系统。尽管大语言模型（LLMs）表现良好，但其高计算成本和延迟限制了大规模部署。为此，我们提出Tool-MCoT，一个基于外部工具增强的思维链训练的小语言模型（SLM），用于内容安全审核。通过在由大模型生成的工具增强型思维链数据上微调，我们证明该小模型能够有效利用工具提升推理与决策能力。实验表明，微调后的小模型取得显著性能提升；同时，模型学会选择性调用工具，在必要时才调用，实现了审核准确率与推理效率的平衡。
原文摘要 · Abstract (English)
The growth of online platforms and user content requires strong content moderation systems that can handle complex inputs from various media types. While large language models (LLMs) are effective, their high computational cost and latency present significant challenges for scalable deployment. To address this, we introduce Tool-MCoT, a small language model (SLM) fine-tuned for content safety moderation leveraging external framework. By training our model on tool-augmented chain-of-thought data generated by LLM, we demonstrate that the SLM can learn to effectively utilize these tools to improve its reasoning and decision-making. Our experiments show that the fine-tuned SLM achieves significant performance gains. Furthermore, we show that the model can learn to use these tools selectively, achieving a balance between moderation accuracy and inference efficiency by calling tools only when necessary.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。