mlsanityData to 14 Oct 2026

Symbolic Reasoning Frameworks Trigger Memory-Mediated Ecosystem Dynamics in Multi-Agent LLM Systems

arXiv 2606.07552 ↗First posted 22 May 2026cs.MA cs.AI cs.LGcited by 0

On the map: LLM reasoning → Reasoning for agents, planning and code

also in 1 more
Abstract (arXiv)

Large language models exhibit a risk-averse "turtle" bias as strategic agents. We show that injecting a symbolic reasoning framework as a per-round reflective prompt into one agent acts as a small perturbation whose consequences are not per-decision but emergent: the agent's risk posture is unchanged in isolation, yet over a campaign of accumulating memory and multi-agent interaction the conditions settle into distinct, condition-associated winner ecosystems. In a 7-player Warring States Diplomacy variant (61 games, 6 conditions), the winner distribution differs sharply across the four primary conditions (41 games; permutation omnibus p approximately 0.001): control -> Yan (7/11); I-Ching yarrow -> Yan/Chu co-dominance with Qin fully suppressed (0/10); Tarot -> Qin (5/10); scrambled-text ablation -> Qi (5/10). The scrambled->Qi attractor is robust (vs. pooled and control alone, p = 0.006 and 0.012); tarot->Qin is denominator-dependent (0.006 pooled, 0.064 vs. control). Han never wins and shows no survival difference (Fisher p = 1.0); neither framework's content predicts actions (chi-squared p = 0.95 hexagram, 0.69 Tarot). A memory-free decision-isolation probe (960 calls) shows the process does not change the agent's risk posture in isolation (Friedman p = 0.45; I-Ching p = 0.60; Tarot perturbs move content but not risk, p = 0.021). A 2x2 factorial separating yarrow's decision-time and learning-time components reveals a non-additive interaction: each alone freezes the board (50-60% stalemates), combined they produce zero (permutation p ~ 5e-5). Testing relocates Qin suppression to rival (Chu) expansion governed by campaign memory depth, not the oracle (p = 0.55). We present this as an observation paper: agent-level framework choice produces distinctive, non-additive system-level consequences, transmitted through emergent memory and multi-agent dynamics, not per-decision effects.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 22 of its 24 references that are in the map. First posted on arXiv 22 May 2026.

Builds on

22 of 24 references.

Closest ancestors · 2

Same direction, newest first

  1. 2026LUDOBENCH: Evaluating LLM Behavioural Decision-Making Through Spot-Based Board Game Scenarios in Ludo7 Apr
  2. 2024Can Machines Think Like Humans? A Behavioral Evaluation of LLM Agents in Dictator Games28 Oct

Other references · 20

  1. 2026Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontierscited by 65
  2. 2023Welfare Diplomacy: Benchmarking Language Model Cooperationcited by 45
  3. 2024Richelieu: Self-Evolving LLM-Based Agents for AI Diplomacycited by 36
  4. 2023The Machine Psychology of Cooperation: Can GPT models operationalise prompts for altruism, cooperation, competitiveness and selfishness in economic games?cited by 22
  5. 2026The Memory Curse: How Expanded Recall Erodes Cooperative Intent in LLM Agentscited by 10
  6. 2025Democratizing Diplomacy: A Harness for Evaluating Any Large Language Model on Full-Press Diplomacycited by 7
  7. 2026Strategic Algorithmic Monoculture: Experimental Evidence from Coordination Gamescited by 7
  8. 2025Do Role-Playing Agents Practice What They Preach? Belief-Behavior Consistency in LLM-Based Simulations of Human Trustcited by 6
  9. 2026Persona Vectors in Games: Measuring and Steering Strategies via Activation Vectorscited by 6
  10. 2025DipLLM: Fine-Tuning LLM for Strategic Decision-making in Diplomacycited by 4
  11. 2025Do Persona-Infused LLMs Affect Performance in a Strategic Reasoning Game?cited by 4
  12. 2025Bias-Adjusted LLM Agents for Human-Like Decision-Making via Behavioral Economicscited by 2
8 more
  1. 2025Large language models replicate and predict human cooperation across experiments in game theorycited by 2
  2. 2025Benevolent Dictators? On LLM Agent Behavior in Dictator Gamescited by 2
  3. 2025Measuring Fine-Grained Negotiation Tactics of Humans and LLMs in Diplomacycited by 2
  4. 2026Training Generalizable Collaborative Agents via Strategic Risk Aversioncited by 2
  5. 2026CivBench: Progress-Based Evaluation for LLMs' Strategic Decision-Making in Civilization Vcited by 2
  6. 2026Coalition Formation in LLM Agent Networks: Stability Analysis and Convergence Guaranteescited by 2
  7. 2026Data-Centric Interpretability for LLM-based Multi-Agent Reinforcement Learningcited by 1
  8. 2026Statistical Properties of the King Wen Sequence: An Anti-Habituation Structure That Does Not Improve Neural Network Trainingcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025The Fragility of Strategic Thinking in Large Language Models
  2. 2026Readable Minds: Emergent Theory-of-Mind-Like Behavior in LLM Poker Agents
  3. 2025Playing games with Large language models: Randomness and strategy
  4. 2026MINDGAMES: A Live Arena for Evaluating Social and Strategic Reasoning in Multi-Agent LLMs
  5. 2026Sparks of Cooperative Reasoning: LLMs as Strategic Hanabi Agents
  6. 2025Limits of Emergent Reasoning of Large Language Models in Agentic Frameworks for Deterministic Games
  7. 2026LLMs Are Not Good Strategists, Yet Memory-Enhanced Agency Boosts Reasoning
  8. 2026Multi-Agent Strategic Games with LLMs