策略，而非收益：标准型博弈的行为嵌入
Strategy, Not Payoffs: A Behavioural Embedding of Normal-Form Games
- University of Waterloo（滑铁卢大学）
机构由 AI 辅助整理，请以论文原文为准。
中文总结 AI 辅助
该研究提出轻量双特征行为嵌入，可预测大型语言模型在不同标准型博弈间微调后的策略能力变化，证明策略能力迁移由决策行为结构而非收益几何决定。
中文摘要 AI 辅助
学习一项策略任务带来的改变远超直接教授的内容：在一个博弈上微调，既可能增强也可能削弱智能体在另一博弈中的推理能力。然而，理解并预测这种策略能力的迁移，仍是大型语言模型（LLM）面临的关键挑战。标准型博弈为分析该现象提供了理想的测试平台，因为它们具有明确界定的收益和特征清晰的均衡行为。本研究探讨博弈嵌入是否能解释并预测LLM在不同博弈间微调后策略能力的变化，提出一种轻量的双特征嵌入，捕捉基本行为需求：纳什均衡的熵以及最优响应对对手行动的敏感性。研究表明，现有已发表的结构嵌入主要是记忆博弈身份，无法泛化；而本研究的行为嵌入能可靠预测在未见过的博弈上的性能变化。这些结果证明，LLM中策略能力的迁移并非由博弈的收益几何决定，而是由其所需的决策行为的底层结构决定。
英文摘要
Learning a strategic task changes more than what is directly taught: fine-tuning on one game can either enhance or degrade an agent's ability to reason in another. Understanding and predicting this transfer of strategic capabilities, however, remains a key challenge for large language models (LLMs). Normal-form games provide an ideal testbed for analyzing this phenomenon, as they feature explicitly defined payoffs and well-characterized equilibrium behaviours. In this work, we investigate whether game embeddings can explain and predict changes in LLM strategic capabilities following fine-tuning across different games. We propose a lightweight two-feature embedding that captures fundamental behavioural demands: the entropy of the Nash equilibrium and the sensitivity of optimal responses to an opponent's action. We show that while existing published structural embeddings primarily memorize game identities and fail to generalize, our behavioural embedding reliably predicts performance changes on held-out games. These results demonstrate that the transfer of strategic capabilities in LLMs is not dictated by the payoff geometry of a game, but by the underlying structure of the decision-making behaviour it requires.
