让大模型学会根据问题难易自动选是否用工具，减少幻觉又省成本。
Adapting While Learning: Grounding LLMs for Scientific Problems with Intelligent Tool Usage Adaptation
- 分两阶段训练：先学科学知识，再根据准确率判断问题难易
- 在6个科学数据集上准确率提升29.11%，工具使用更精准
- 适合需要高可靠性的科研场景，尤其关注成本与准确率平衡
大型语言模型在解决科学问题上表现优异，但常出现幻觉。虽可通过集成工具缓解，但过度依赖工具的微调模型会增加不必要开销。受人类专家评估问题复杂度后选择策略的启发，我们提出一种两组件微调方法——适配式学习（AWL）。第一部分世界知识学习（WKL）使模型通过学习工具生成的解法内化科学知识；第二部分工具使用适应（TUA）根据模型准确率将问题分类为简单或困难，训练模型对简单问题直接推理，对困难问题调用工具。我们在气候科学、流行病学、物理学等多个领域共六个基准数据集上验证该方法。相比原始指令模型（8B），经AWL微调后的模型答案准确率提升29.11%，工具使用准确率提高12.72%，甚至在四个自建数据集上超越GPT-4o和Claude-3.5。代码已开源于https://github.com/Rose-STL-Lab/Adapting-While-Learning。
原文摘要 · Abstract (English)
Large Language Models (LLMs) demonstrate promising capabilities in solving scientific problems but often suffer from the issue of hallucination. While integrating LLMs with tools can mitigate this issue, models fine-tuned on tool usage become overreliant on them and incur unnecessary costs. Inspired by how human experts assess problem complexity before selecting solutions, we propose a novel two-component fine-tuning method, Adapting While Learning (AWL). In the first component, World Knowledge Learning (WKL), LLMs internalize scientific knowledge by learning from tool-generated solutions. In the second component, Tool Usage Adaptation (TUA), we categorize problems as easy or hard based on the model's accuracy, and train it to maintain direct reasoning for easy problems while switching to tools for hard ones. We validate our method on six scientific benchmark datasets across climate science, epidemiology, physics, and other domains. Compared to the original instruct model (8B), models post-trained with AWL achieve 29.11% higher answer accuracy and 12.72% better tool usage accuracy, even surpassing state-of-the-art models including GPT-4o and Claude-3.5 on four custom-created datasets. Our code is open-source at https://github.com/Rose-STL-Lab/Adapting-While-Learning.
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。