强化智能体在科学构思中的创造性：以“夜科学”为鉴
Reinforcing Agentic Creativity in Scientific Ideation with Night Science
浏览论文内容
中文总结 AI 辅助
提出AI Night-Scientist框架，用强化学习训练LLM在科学构思中平衡探索与利用，显著提升提案多样性和原创性，表明创造力可学习且需语义指导。
中文摘要 AI 辅助
大型语言模型（LLMs）在结构化、可验证的任务上表现出色，但其低熵偏差可能导致输出同质化且可预测，限制了它们在开放式科学构思中的实用性。然而，有效的科学发现涵盖更广泛的创造性光谱：从结构化的“日科学”到松散结构、偶然发现的“夜科学”，后者能触及通常不被考虑的想法。我们引入了AI夜科学家（AI Night-Scientist），一个基于强化学习的智能体框架，教会模型何时以及如何偏离可预测的推理。基于认知科学，我们沿三个轴对创造力进行建模：行动（做什么以及如何创造性地做）、过程（何时探索与利用）和结果（所产生想法的新颖性和实用性）。我们利用这些轴通过GRPO训练模型，在整个训练过程中让模型接触不同程度和形式的创造力。这产生了更多样化的科学提案，将研究方向的范围扩大了27.8%，贡献类型扩大了14.9%，相对于基础模型。它还提高了预测的引用影响，最高达32.0个百分点，以及原创性，提高了66.2点。这些增益无法通过简单提高解码温度来复现；相反，我们发现指定追求何种创造力的语义指导至关重要。总体而言，我们的结果表明，创造力是一种可学习的、多层面的能力，可以被塑造以帮助研究人员触及那些通常不被LLMs探索的想法。
英文摘要
Large language models (LLMs) excel at structured, verifiable tasks, but their low-entropy bias can produce homogeneous and predictable outputs, limiting their utility for open-ended scientific ideation. Effective discovery, however, spans a broader creative spectrum: from structured day science to loosely structured, serendipitous night science that reaches ideas beyond those typically considered. We introduce AI Night-Scientist, an agentic framework that uses reinforcement learning to teach models when and how to depart from predictable reasoning. Grounded in cognitive science, we model creativity along three axes: action (what to do and how creatively), process (when to explore versus exploit), and outcome (the novelty and usefulness of the resulting idea). We use these axes to train models with GRPO, exposing them to varying degrees and forms of creativity throughout training. This produces substantially more diverse scientific proposals, expanding the range of research directions by 27.8% and contribution types by 14.9% over the base model. It also improves predicted citation impact by up to 32.0 percentage points and originality by 66.2 points. These gains cannot be reproduced by simply increasing decoding temperature; instead, we find that semantic guidance specifying what kind of creativity to pursue is critical. Overall, our results suggest that creativity is a learnable, multi-level ability that can be shaped to help researchers reach ideas beyond those typically explored by LLMs.
发表机构
- University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
- Microsoft（微软）
- Microsoft Research（微软研究院）
机构由 AI 辅助整理，请以论文原文为准。
