Stratagem: Learning Transferable Reasoning via Trajectory-Modulated Game Self-Play

• URL: http://arxiv.org/abs/2604.17696v1
• Date: Mon, 20 Apr 2026 01:20:31 GMT
• Title: Stratagem: Learning Transferable Reasoning via Trajectory-Modulated Game Self-Play
• Abstract summary: We present STRATAGEM, which addresses two fundamental barriers to reasoning transfer: domain specificity and contextual stasis.<n>Experiments across mathematical reasoning, general reasoning, and code generation benchmarks demonstrate substantial improvements.
• Score: 85.2683525068614
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Games offer a compelling paradigm for developing general reasoning capabilities in language models, as they naturally demand strategic planning, probabilistic inference, and adaptive decision-making. However, existing self-play approaches rely solely on terminal game outcomes, providing no mechanism to distinguish transferable reasoning patterns from game-specific heuristics. We present STRATAGEM, which addresses two fundamental barriers to reasoning transfer: domain specificity, where learned patterns remain anchored in game semantics, and contextual stasis, where static game contexts fail to cultivate progressive reasoning. STRATAGEM selectively reinforces trajectories exhibiting abstract, domain-agnostic reasoning through a Reasoning Transferability Coefficient, while incentivizing adaptive reasoning development via a Reasoning Evolution Reward. Experiments across mathematical reasoning, general reasoning, and code generation benchmarks demonstrate substantial improvements, with particularly strong gains on competition-level mathematics where multi-step reasoning is critical. Ablation studies and human evaluation confirm that both components contribute to transferable reasoning.

Related papers

• Spatial Reasoning via Modality Switching Between Language and Symbolic Representation [26.746461996486474]
  This research investigates whether grounding multi-hop textual-spatial stories into geometry-aware modalities improves reasoning.<n>We introduce a switching metric based on trustworthiness and complexity signals, which estimates when grounding a spatial story into structure is likely to improve performance.<n>Across our settings, switching from natural language-based reasoning to a grid-based representation improves LLM performance by up to 42%, highlighting the importance of modality choice in shaping reasoning outcomes.
  arXiv  Detail & Related papers  (2026-06-30T08:02:07Z)
• Look on Demand: A Cognitive Scheduling Framework for Visual Evidence Acquisition in Multimodal Reasoning [82.86343313807158]
  We argue that a central challenge is how and when visual evidence is introduced into the reasoning process.<n>Motivated by this insight, we propose CSMR, a multimodal reasoning framework in which a language model controls the reasoning process.
  arXiv  Detail & Related papers  (2026-05-27T08:43:13Z)
• Learning Structured Reasoning via Tractable Trajectory Control [99.75278337895024]
  Ctrl-R is a framework for learning structured reasoning via tractable trajectory control.<n>We show that Ctrl-R enables effective exploration and internalization of previously unattainable reasoning patterns.
  arXiv  Detail & Related papers  (2026-03-02T09:18:19Z)
• GeoReason: Aligning Thinking And Answering In Remote Sensing Vision-Language Models Via Logical Consistency Reinforcement Learning [12.987952829880363]
  GeoReason is a framework designed to synchronize internal thinking with final decisions.<n>We first construct GeoReason-Bench, a logic-driven dataset containing 4,000 reasoning trajectories.<n>We then formulate a two-stage training strategy: (1) Supervised Knowledge Initialization to equip the model with reasoning syntax and domain expertise, and (2) Consistency-Aware Reinforcement Learning to refine deductive reliability.
  arXiv  Detail & Related papers  (2026-01-07T17:26:41Z)
• From Hypothesis to Premises: LLM-based Backward Logical Reasoning with Selective Symbolic Translation [8.104087344683604]
  We propose a novel framework, Hypothesis-driven Backward Logical Reasoning (HBLR)<n>The core idea is to integrate confidence-aware symbolic translation with hypothesis-driven backward reasoning.<n>HBLR consistently outperforms strong baselines in both accuracy and efficiency.
  arXiv  Detail & Related papers  (2025-12-03T01:52:31Z)
• LLMs as Strategic Agents: Beliefs, Best Response Behavior, and Emergent Heuristics [0.0]
  Large Language Models (LLMs) are increasingly applied to domains that require reasoning about other agents' behavior.<n>We show that current frontier models exhibit belief-coherent best-response behavior at targeted reasoning memorization.<n>Under increasing complexity, explicit recursion gives way to internally generated rules of choice that are stable, model-specific, and distinct from known human biases.
  arXiv  Detail & Related papers  (2025-10-12T21:40:29Z)
• CTRLS: Chain-of-Thought Reasoning via Latent State-Transition [57.51370433303236]
  Chain-of-thought (CoT) reasoning enables large language models to break down complex problems into interpretable intermediate steps.<n>We introduce groundingS, a framework that formulates CoT reasoning as a Markov decision process (MDP) with latent state transitions.<n>We show improvements in reasoning accuracy, diversity, and exploration efficiency across benchmark reasoning tasks.
  arXiv  Detail & Related papers  (2025-07-10T21:32:18Z)
• A Survey on Latent Reasoning [100.54120559169735]
  Large Language Models (LLMs) have demonstrated impressive reasoning capabilities.<n>CoT reasoning that verbalizes intermediate steps limits the model's expressive bandwidth.<n>Latent reasoning tackles this bottleneck by performing multi-step inference entirely in the model's continuous hidden state.
  arXiv  Detail & Related papers  (2025-07-08T17:29:07Z)
• AdapThink: Adaptive Thinking Preferences for Reasoning Language Model [32.47427081297578]
  Reinforcement Learning (RL)-based post-training has significantly advanced the complex reasoning capabilities of language models.<n>However, this slow thinking'' paradigm presents a critical challenge to reasoning efficiency.<n>We propose AdapThink, an adaptive post-training framework designed to induce more efficient thinking.
  arXiv  Detail & Related papers  (2025-06-23T02:06:04Z)
• From Heuristic to Analytic: Cognitively Motivated Strategies for Coherent Physical Commonsense Reasoning [66.98861219674039]
  Heuristic-Analytic Reasoning (HAR) strategies drastically improve the coherence of rationalizations for model decisions. Our findings suggest that human-like reasoning strategies can effectively improve the coherence and reliability of PLM reasoning.
  arXiv  Detail & Related papers  (2023-10-24T19:46:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.