论文/arXiv 2610.08781·22 天前

IdeaAnchor：教 LLMs 将文献转化为研究想法

提出 IdeaAnchor，用结构化规格监督 LLM 从相关论文综合研究想法，并在研究构思实验中验证质量提升。

AI 整理

• 数据与采集
• 具身模型
• 端侧部署

原文链接：IdeaAnchor: Teaching LLMs to Turn Literature into Research Ideas（arXiv 2610.08781 · 发布日期：2026-10-06）

IdeaAnchor: Teaching LLMs to Turn Literature into Research Ideas

提出 IdeaAnchor，用结构化规格监督 LLM 从相关论文综合研究想法，并在研究构思实验中验证质量提升。

核心要点

• 研究构思需综合相关论文发现空白，但提示或反馈方法缺少论文合成方式的结构化监督。
• IdeaAnchor 将每篇输入论文的功能角色、关系和目标合成标准编码为特权信号。
• 该方法从已发表论文中挖掘实例，并结合示范学习、自蒸馏、强化学习与推理时检索。
• 实验显示构思质量持续提升，分析表明锚点训练强化创造性合成，检索增强细节展开。

验证范围与局限：验证范围为基于已发表论文挖掘实例的研究构思生成实验。

英文摘要 (Abstract)

Scientific research often begins by synthesizing ideas from a set of related papers to identify gaps and formulate new directions. However, training language models to perform this form of literature-grounded ideation remains challenging, as existing approaches based on prompting or feedback lack structured supervision for how papers should be synthesized. We introduce IdeaAnchor, a paradigm for training LLMs to perform research ideation using structured specifications as privileged signals. Each IdeaAnchor instance encodes how each input paper should be synthesized into a successful idea, including their functional roles, relationships, and target synthesis criteria. We build this paradigm by mining instances from published papers, capturing how real ideas emerge from prior literature. We then train models via demonstration, self-distillation, and reinforcement learning, and further enhance generation with retrieval at inference time. Experiments show consistent improvements in ideation quality. Our analysis reveals a functional decomposition: anchor-based training strengthens creative synthesis, retrieval enhances detail elaboration, and combining both yields the best performance.

作者：Ziyu Chen, Yilun Zhao, Jiashuo Sun, Yiling Ma, Manasi S. Patwardhan, Arman Cohan

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.08781：IdeaAnchor: Teaching LLMs to Turn Literature into Research Ideas访问 2026-10-21