arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.00482cs.CL
收藏

面向微领域中特定领域大语言模型的基于蒸馏的智能体微调

Agent Fine-tuning through Distillation for Domain-specific LLMs in Microdomains

Research(研究);Development Group Hitachi, Ltd. Tokyo, Japan(开发组 日立株式会社 东京日本)

Yawen Xue, Masaya Tsunokake, Yuta Koreeda, Ekant Muljibhai Amin, Takashi Sumiyoshi, Yasuhiro Sogawa

AI总结:

针对通用智能体大语言模型在专业技术微领域效果不明的问题,提出结合领域专属数据与自蒸馏推理轨迹的智能体微调方法,搭配检索增强提示和上下文-答案提取器,在JP1认证考题上较基线提升14%性能。

摘要:

智能体大语言模型(LLM)因能自主与外部环境交互并执行多步推理任务而备受关注。大多数方法通过少样本提示的上下文学习来利用这些能力,但这往往导致输入冗长且计算成本更高。智能体微调提供了另一种思路:通过在相关数据和演示轨迹上训练,让大语言模型内化程序推理和特定领域知识。此前的研究聚焦于通用领域,但其在专业技术微领域中的效果仍不明确。本文在日立JP1中间件(一个专业IT运维微领域)中探索用于领域适配的智能体微调。我们利用从领域手册中提取的JP1专属数据集、以及由大语言模型自身生成的蒸馏推理轨迹来微调大语言模型,提升了决策准确率和搜索效率。推理阶段,我们使用结合检索增强生成的智能体提示,并引入上下文-答案提取器以提升信息相关性。在JP1认证考试题目上,我们的方法相比基线模型实现了14%的性能提升,证明了智能体微调在复杂微领域中用于特定领域推理的潜力。

英文摘要:

Agentic large language models (LLMs) have become prominent for autonomously interacting with external environments and performing multi-step reasoning tasks. Most approaches leverage these capabilities via in-context learning with few-shot prompts, but this often results in lengthy inputs and higher computational costs. Agent fine-tuning offers an alternative by enabling LLMs to internalize procedural reasoning and domain-specific knowledge through training on relevant data and demonstration trajectories. While prior studies have focused on general domains, their effectiveness in specialized technical microdomains remains unclear. This paper explores agent fine-tuning for domain adaptation within Hitachi's JP1 middleware, a microdomain for specialized IT operations. We fine-tuned LLMs using JP1-specific datasets derived from domain manuals and distilled reasoning trajectories generated by LLMs themselves, enhancing decision making accuracy and search efficiency. During inference, we used an agentic prompt with retrieval-augmented generation and introduced a context-answer extractor to improve information relevance. On JP1 certification exam questions, our method achieved a 14% performance improvement over the base model, demonstrating the potential of agent fine-tuning for domain-specific reasoning in complex microdomains.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑