Skip to content

SAGE: Structured Strategic Reasoning for Efficient LLM Game Playing

Sep 2026 · arXiv.org · Vol abs/2609.34342 · 0 citations · 49 references
Computer Science

TL;DR

SAGE is a training-free inference-time framework that structures LLM strategic reasoning around three coordinated operations: anchor, adapt, and recalibrate, and distills strategically related interactions into counterfactual hypotheses about previously missing considerations, allowing past experience to recalibrate the model's reasoning.

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.

View source

Similar papers

#artificial intelligence Open access Sep 2026

Recursive Reasoning or Statistical Extrapolation? In-Context Learning in Multi-Agent Interdependent Decision-Making

This work extends the mechanistic study of ICL to strategic multi-agent settings, introduces REE as a diagnostic tool for distinguishing reasoning from extrapolation, and provides a reusable framework for probing the boundaries of LLM reasoning in recursive belief tasks.

Y. Liu, Wen-Wen Li, Yi-Fan Dou et al. · 0 citations
Preprint Aug 2026

Level-k Distinguishable Mechanisms for Evaluating Bounded Rationality in LLMs

This work formalises a necessary level-K distinguishability condition for strategic depth inference and builds a suite of novel game structures that meet this standard, and finds that model models maintain accurate strategic depth under recursive reasoning, with strong internal consistency between stated reasoning and...

Binchi Zhang, Atrisha Sarkar · 0 citations
#artificial intelligence Open access Oct 2026

Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict

Evaluating four agents and three model-level interventions, it is found that higher task accuracy does not necessarily correspond to greater epistemic humility: some high-accuracy configurations recognize conflicts during execution but do not communicate unresolved uncertainty in their incorrect final answers.

Kaiser Sun, Bernal Jimenez Gutierrez, Hong-Jun Liu et al. · 1 citation
#artificial intelligence Preprint Sep 2026

OverForge: Reasoning Through Strategies and Tactics Helps Cooperative Lifelong Adaptation

Cooperative language-model agents must coordinate over long horizons and adapt to changing environments and to partners with unfamiliar conventions, yet existing agents map observations to actions without separating persistent coordination strategies from their tactical execution. We introduce OverForge, a training-fre...

Oana Madalina Fron, Ojas Shirekar, Chirag Raman · 0 citations
Preprint Sep 2026

Cheap Talk Stabilizes Strategic Interaction in LLM Agents

Large language models are increasingly deployed as interacting agents, making the persistence of their action policies across repeated interaction critical for reliable multi-agent operation. We investigate whether and how agent-generated, non-binding pre-play communication ("cheap talk") increases such persistence in...

Nunzio Lorè, Hon-Gan Zhu, Babak Heydari · 0 citations
#artificial intelligence Preprint Sep 2026

MARBO: Relational Belief Grounding for LLM Agents in Social Deduction Games

This work introduces Multi-Agent Relational Belief Optimization (MARBO), a belief-grounded preference optimization framework that leverages relational beliefs to guide strategic decisions and in-game speech and enables compact LLM agents to consistently outperform existing baselines.

Yechan Hwang, S. Bae, Jeongmo Kim et al. · 0 citations

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.