Large Language Models (LLMs) often struggle with generating truly innovative ideas, typically defaulting to high-probability, familiar concepts within their training data's "gravity wells." While advanced search-based methods like Tree of Thoughts (ToT) attempt to mitigate this, they are fundamentally limited by their reliance on unprincipled, inconsistent self-evaluation heuristics to guide exploration. To address this gap, we introduce \textbf{Magellan}, a novel framework that reframes creative generation as a principled, guided exploration of an LLM's latent conceptual space. At its core, Magellan employs Monte Carlo Tree Search (MCTS) governed by a hierarchical guidance system. For long-range direction, a "semantic compass" vector, formulated via orthogonal projection, steers the search towards relevant novelty. For local, step-by-step decisions, a landscape-aware value function replaces flawed self-evaluation with an explicit reward structure that balances intrinsic coherence, extrinsic novelty, and narrative progress. Extensive experiments demonstrate that Magellan significantly outperforms strong baselines, including ReAct and ToT, in generating scientific ideas with superior plausibility and innovation. Our work shows that for creative discovery, a principled, guided search is more effective than unconstrained agency, paving the way for LLMs to become more capable partners in innovation.

翻译：大型语言模型（LLM）在生成真正创新性想法时常常遇到困难，通常倾向于回退到其训练数据“引力阱”内的高概率、熟悉概念。尽管基于搜索的先进方法（如思维树（ToT））试图缓解这一问题，但它们从根本上受限于对无原则、不一致的自我评估启发式方法的依赖来引导探索。为弥补这一不足，我们提出了 **Magellan**，这是一个新颖的框架，它将创造性生成重新定义为对LLM潜在概念空间的一种原则性、引导式探索。Magellan的核心是采用由分层引导系统控制的蒙特卡洛树搜索（MCTS）。在长程方向上，一个通过正交投影构建的“语义罗盘”向量引导搜索朝向相关的新颖性。在局部、逐步决策层面，一个具备景观感知的价值函数取代了有缺陷的自我评估，采用了一种显式的奖励结构，平衡内在连贯性、外在新颖性和叙事进展。大量实验表明，在生成具有更优合理性与创新性的科学想法方面，Magellan显著优于包括ReAct和ToT在内的强基线方法。我们的工作表明，对于创造性发现，一种原则性、引导式的搜索比无约束的自主性更为有效，这为LLM成为更具能力的创新合作伙伴铺平了道路。