Papers for

interactive simulation designers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

New method improves large language model game playing strategy and efficiency

SAGE: Structured Strategic Reasoning for Efficient LLM Game Playing

Abstract: A strong LLM strategic agent should reason prospectively over uncertain futures, adapt its strategy to opponents' behavioral tendencies, and continuously recalibrate its decision process from interaction experience. However, incorporating these sources in free-form reasoning could lead to unsupported strategic assumptions, inconsistent opponent estimates, and harmful interference from irrelevant historical interactions. To address these issues, we propose SAGE, a training-free inference-time framework that structures LLM strategic reasoning around three coordinated operations: anchor, adapt, and recalibrate. SAGE first anchors reasoning to an equilibrium policy that provides a strategically valid prior. It then conditions deviations from this anchor on a soft belief over opponent behavioral tendencies, enabling opponent-specific exploitation. Finally, SAGE distills strategically related interactions into counterfactual hypotheses about previously missing considerations, allowing past experience to recalibrate the model's reasoning. We evaluate SAGE on three repeated imperfect-information games: Leduc Hold'em, Liar's Dice, and Goofspiel, against various opponent types in each game. Compared with reasoning-intensive LLM agents, including Suspicion-Agent, ReTA, Agent-Pro, EMO, and Hypothetical Minds, SAGE achieves up to a 127.6% payoff improvement in Liar's Dice while reducing input and output token usage by up to 80% and 90%, respectively. In direct match-up play, it attains non-negative mean payoff against 5/10, 8/10, and 8/10 evaluated opponents in Leduc Hold'em, Liar's Dice, and Goofspiel, respectively, while using relatively fewer tokens. Code is available at https://github.com/chenzhwsysu57/SAGE.

Mon 28 SeptArtificial Intelligence
The gist
Playing games well requires thinking ahead, adapting to how others behave, and learning from past moves. The authors developed a method called SAGE that helps language models do exactly this without extra training. SAGE relies on a solid baseline strategy, adjusts based on what it learns about opponents, and updates its reasoning from past game interactions. This approach improves game performance notably while using fewer computing resources.
Open → 2609.34342v1