arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.26322cs.CL
收藏

SCRIBE:基于工具调用的交互行为推理的结构化链推理

SCRIBE: Structured Chain Reasoning for Interactive Behaviour Explanations using Tool Calling

EPFL(苏黎世联邦理工学院)

Fares Fawzi, Vinitra Swamy, Dominik Glandorf, Tanya Nazaretsky, Tanja Käser

摘要:

语言模型可用于在教育环境中提供交互式的、个性化的学生反馈。然而,现实世界的部署面临着三个关键挑战:隐私问题,有限的计算资源,以及需要教学有效的反应。这些限制需要小型的开源模型,这些模型可以在本地运行,并以正确的信息可靠地支持其输出。我们介绍SCRIBE,多跳,工具增强推理的框架,旨在生成有效的反应,学生的反馈报告的问题。SCRIBE将特定于领域的工具与支持迭代推理、工具使用和错误恢复的自反射推理管道相结合。我们通过两阶段LoRA对合成GPT-4 o生成的数据进行微调,将这些功能扩展到3B和8B模型中。与人类对齐的GPT-Judge评估和108名学生的用户研究表明,8B-SCRIBE模型在相关性和可操作性等关键维度上达到了与更大模型相当或更高的质量,同时被学生认为与GPT-4 o和Llama-3.3 70 B相当。这些发现证明了SCRIBE在低资源、隐私敏感的教育应用中的可行性。

英文摘要:

Language models can be used to provide interactive, personalized student feedback in educational settings. However, real-world deployment faces three key challenges: privacy concerns, limited computational resources, and the need for pedagogically valid responses. These constraints require small, open-source models that can run locally and reliably ground their outputs in correct information. We introduce SCRIBE, a framework for multi-hop, tool-augmented reasoning designed to generate valid responses to student questions about feedback reports. SCRIBE combines domain-specific tools with a self-reflective inference pipeline that supports iterative reasoning, tool use, and error recovery. We distil these capabilities into 3B and 8B models via two-stage LoRA fine-tuning on synthetic GPT-4o-generated data. Evaluation with a human-aligned GPT-Judge and a user study with 108 students shows that 8B-SCRIBE models achieve comparable or superior quality to much larger models in key dimensions such as relevance and actionability, while being perceived on par with GPT-4o and Llama-3.3 70B by students. These findings demonstrate the viability of SCRIBE for low-resource, privacy-sensitive educational applications.

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。

↑