Most machine learning approaches to scientific discovery frame hypotheses as end-to-end predictions, obscuring the incremental structure of scientific reasoning. We propose The Hypothesis Game, a symbolic formalism for hypothesis refinement in which LLM agents operate on a shared hypothesis state using a fixed grammar of reasoning moves. The framework is motivated by the observation that scientific progress often proceeds through small, localized revisions, grounded in domain context, rather than extensive rewrites. We instantiate a minimal game with LLM agents and evaluate it on pathway-level mechanistic refinement tasks. In the primary setting of corruption recovery, where hypotheses contain controlled errors, the game-based approach consistently removes more errors and achieves higher precision than strong prompting baselines, while preserving valid structure through incremental edits. In a secondary reconstruction setting from partial cues, it performs comparably to the strongest baseline, indicating that explicit move-based refinement remains competitive even when ground-truth recovery is difficult. These findings support game-based reasoning as a principled route to more controllable, interpretable, and transferable hypothesis refinement systems for scientific discovery.

翻译：大多数用于科学发现的机器学习方法将假设构建为端到端的预测，这模糊了科学推理的增量结构。我们提出了“假设游戏”，一种用于假设精炼的符号形式化框架，其中LLM智能体使用固定的推理移动语法对共享的假设状态进行操作。该框架的动机源于观察到科学进步通常通过基于领域背景的、局部的小幅修订实现，而非大规模的重写。我们使用LLM智能体实例化了一个最小化游戏，并在通路层面的机制精炼任务上对其进行了评估。在主要设定——错误恢复（假设包含受控错误）中，基于游戏的方法持续移除了更多错误，并实现了比强提示基线更高的精确度，同时通过增量编辑保留了有效结构。在基于部分线索的次要重建设定中，其表现与最强基线相当，这表明即使在地面真值恢复困难的情况下，基于显式移动的精炼方法仍具有竞争力。这些发现支持基于游戏的推理作为一种原则性路径，可用于构建更可控、可解释和可迁移的科学发现假设精炼系统。