We report a structural convergence among four influential theories of mind: Kahneman's dual-system theory, Friston's predictive processing, Minsky's society of mind, and Clark's extended mind, emerging unintentionally within a practical AI architecture known as Agentic Flow. Designed to address the limitations of large language models (LLMs), Agentic Flow comprises five interlocking modules: Retrieval, Cognition, Control, Memory, and Action, organized into a repeatable cognitive loop. Although originally inspired only by Minsky and Clark, subsequent analysis revealed that its structure echoes computational motifs from all four theories, suggesting that theoretical convergence can emerge naturally from implementation demands rather than deliberate synthesis. Controlled evaluations confirmed this: the structured agent achieved 95.8% task success versus 62.3% for baseline LLMs, demonstrating robust constraint adherence and reproducible reasoning. We describe this convergence under a broader descriptive meta-architecture called PEACE, highlighting recurring design patterns such as predictive modeling, associative recall, and error-sensitive control. Later formalized as the Structured Cognitive Loop (SCL), this framework generalizes the same principles as a foundation for behavioral intelligence in LLM-based agents. Rather than claiming theoretical unification, this paper proposes that intelligent architectures may evolve toward shared structural patterns shaped by practical constraints. As a position paper, it aims to frame this convergence as an interpretive reflection rather than a finalized theory, inviting further theoretical and experimental dialogue. Agentic Flow, or equivalently the Structured Cognitive Loop, thus offers a glimpse of how a unified cognitive form can arise not from abstraction, but from the necessities of real-world reasoning.

Adversarial optimization algorithms that explicitly search for flaws in agents' policies have been successfully applied to finding robust and diverse policies in multi-agent settings. However, the success of adversarial optimization has been largely limited to zero-sum settings because its naive application in cooperative settings leads to a critical failure mode: agents are irrationally incentivized to self-sabotage, blocking the completion of tasks and halting further learning. To address this, we introduce Rationality-preserving Policy Optimization (RPO), a formalism for adversarial optimization that avoids self-sabotage by ensuring agents remain rational--that is, their policies are optimal with respect to some possible partner policy. To solve RPO, we develop Rational Policy Gradient (RPG), which trains agents to maximize their own reward in a modified version of the original game in which we use opponent shaping techniques to optimize the adversarial objective. RPG enables us to extend a variety of existing adversarial optimization algorithms that, no longer subject to the limitations of self-sabotage, can find adversarial examples, improve robustness and adaptability, and learn diverse policies. We empirically validate that our approach achieves strong performance in several popular cooperative and general-sum environments. Our project page can be found at https://rational-policy-gradient.github.io.

Multi-Agent Systems (MAS) have emerged as a powerful paradigm for modeling complex interactions among autonomous entities in distributed environments. In Multi-Agent Reinforcement Learning (MARL), communication enables coordination but can lead to inefficient information exchange, since agents may generate redundant or non-essential messages. While prior work has focused on boosting task performance with information exchange, the existing research lacks a thorough investigation of both the appropriate definition and the optimization of communication protocols (communication topology and message). To fill this gap, we introduce a generalized framework for learning multi-round communication protocols that are both effective and efficient. Within this framework, we propose three novel Communication Efficiency Metrics (CEMs) to guide and evaluate the learning process: the Information Entropy Efficiency Index (IEI) and Specialization Efficiency Index (SEI) for efficiency-augmented optimization, and the Topology Efficiency Index (TEI) for explicit evaluation. We integrate IEI and SEI as the adjusted loss functions to promote informative messaging and role specialization, while using TEI to quantify the trade-off between communication volume and task performance. Through comprehensive experiments, we demonstrate that our learned communication protocol can significantly enhance communication efficiency and achieves better cooperation performance with improved success rates.

Multi-agent multi-objective systems (MAMOS) have emerged as powerful frameworks for modelling complex decision-making problems across various real-world domains, such as robotic exploration, autonomous traffic management, and sensor network optimisation. MAMOS offers enhanced scalability and robustness through decentralised control and more accurately reflects inherent trade-offs between conflicting objectives. In MAMOS, each agent uses utility functions that map return vectors to scalar values. Existing MAMOS optimisation methods face challenges in handling heterogeneous objective and utility function settings, where training non-stationarity is intensified due to private utility functions and the associated policies. In this paper, we first theoretically prove that direct access to, or structured modeling of, global utility functions is necessary for the Bayesian Nash Equilibrium under decentralised execution constraints. To access the global utility functions while preserving the decentralised execution, we propose an Agent-Attention Multi-Agent Multi-Objective Reinforcement Learning (AA-MAMORL) framework. Our approach implicitly learns a joint belief over other agents' utility functions and their associated policies during centralised training, effectively mapping global states and utilities to each agent's policy. In execution, each agent independently selects actions based on local observations and its private utility function to approximate a BNE, without relying on inter-agent communication. We conduct comprehensive experiments in both a custom-designed MAMO Particle environment and the standard MOMALand benchmark. The results demonstrate that access to global preferences and our proposed AA-MAMORL significantly improve performance and consistently outperform state-of-the-art methods.

In this paper, we propose capturing and utilizing \textit{Temporal Information through Graph-based Embeddings and Representations} or \textbf{TIGER} to enhance multi-agent reinforcement learning (MARL). We explicitly model how inter-agent coordination structures evolve over time. While most MARL approaches rely on static or per-step relational graphs, they overlook the temporal evolution of interactions that naturally arise as agents adapt, move, or reorganize cooperation strategies. Capturing such evolving dependencies is key to achieving robust and adaptive coordination. To this end, TIGER constructs dynamic temporal graphs of MARL agents, connecting their current and historical interactions. It then employs a temporal attention-based encoder to aggregate information across these structural and temporal neighborhoods, yielding time-aware agent embeddings that guide cooperative policy learning. Through extensive experiments on two coordination-intensive benchmarks, we show that TIGER consistently outperforms diverse value-decomposition and graph-based MARL baselines in task performance and sample efficiency. Furthermore, we conduct comprehensive ablation studies to isolate the impact of key design parameters in TIGER, revealing how structural and temporal factors can jointly shape effective policy learning in MARL. All codes can be found here: https://github.com/Nikunj-Gupta/tiger-marl.

In graph-structured multi-agent reinforcement learning (MARL) adversarial tasks such as pursuit and confrontation, agents must coordinate under highly dynamic interactions, where sparse rewards hinder efficient policy learning. We propose Adaptive Regularized Multi-Agent Soft Actor-Critic (ARAC), which integrates an attention-based graph neural network (GNN) for modeling agent dependencies with an adaptive divergence regularization mechanism. The GNN enables expressive representation of spatial relations and state features in graph environments. Divergence regularization can serve as policy guidance to alleviate the sparse reward problem, but it may lead to suboptimal convergence when the reference policy itself is imperfect. The adaptive divergence regularization mechanism enables the framework to exploit reference policies for efficient exploration in the early stages, while gradually reducing reliance on them as training progresses to avoid inheriting their limitations. Experiments in pursuit and confrontation scenarios demonstrate that ARAC achieves faster convergence, higher final success rates, and stronger scalability across varying numbers of agents compared with MARL baselines, highlighting its effectiveness in complex graph-structured environments.

Multi-agent reinforcement learning (MARL) has demonstrated remarkable performance in multi-agent collaboration problems and has become a prominent topic in artificial intelligence research in recent years. However, traditional credit assignment schemes in MARL cannot reliably capture individual contributions in strongly coupled tasks while maintaining training stability, which leads to limited generalization capabilities and hinders algorithm performance. To address these challenges, we propose a Historical Interaction-Enhanced Shapley Policy Gradient Algorithm (HIS) for Multi-Agent Credit Assignment, which employs a hybrid credit assignment mechanism to balance base rewards with individual contribution incentives. By utilizing historical interaction data to calculate the Shapley value in a sample-efficient manner, HIS enhances the agent's ability to perceive its own contribution, while retaining the global reward to maintain training stability. Additionally, we provide theoretical guarantees for the hybrid credit assignment mechanism, ensuring that the assignment results it generates are both efficient and stable. We evaluate the proposed algorithm in three widely used continuous-action benchmark environments: Multi-Agent Particle Environment, Multi-Agent Mu-JoCo, and Bi-DexHands. Experimental results demonstrate that HIS outperforms state-of-the-art methods, particularly excelling in strongly coupled, complex collaborative tasks.

This dissertation explores the application of multi-agent reinforcement learning (MARL) for handling deadlocks in intralogistics systems that rely on autonomous mobile robots (AMRs). AMRs enhance operational flexibility but also increase the risk of deadlocks, which degrade system throughput and reliability. Existing approaches often neglect deadlock handling in the planning phase and rely on rigid control rules that cannot adapt to dynamic operational conditions. To address these shortcomings, this work develops a structured methodology for integrating MARL into logistics planning and operational control. It introduces reference models that explicitly consider deadlock-capable multi-agent pathfinding (MAPF) problems, enabling systematic evaluation of MARL strategies. Using grid-based environments and an external simulation software, the study compares traditional deadlock handling strategies with MARL-based solutions, focusing on PPO and IMPALA algorithms under different training and execution modes. Findings reveal that MARL-based strategies, particularly when combined with centralized training and decentralized execution (CTDE), outperform rule-based methods in complex, congested environments. In simpler environments or those with ample spatial freedom, rule-based methods remain competitive due to their lower computational demands. These results highlight that MARL provides a flexible and scalable solution for deadlock handling in dynamic intralogistics scenarios, but requires careful tailoring to the operational context.

Monte Carlo Tree Search (MCTS), which leverages Upper Confidence Bound for Trees (UCTs) to balance exploration and exploitation through randomized sampling, is instrumental to solving complex planning problems. However, for multi-agent planning, MCTS is confronted with a large combinatorial action space that often grows exponentially with the number of agents. As a result, the branching factor of MCTS during tree expansion also increases exponentially, making it very difficult to efficiently explore and exploit during tree search. To this end, we propose MALinZero, a new approach to leverage low-dimensional representational structures on joint-action returns and enable efficient MCTS in complex multi-agent planning. Our solution can be viewed as projecting the joint-action returns into the low-dimensional space representable using a contextual linear bandit problem formulation. We solve the contextual linear bandit problem with convex and $μ$-smooth loss functions -- in order to place more importance on better joint actions and mitigate potential representational limitations -- and derive a linear Upper Confidence Bound applied to trees (LinUCT) to enable novel multi-agent exploration and exploitation in the low-dimensional space. We analyze the regret of MALinZero for low-dimensional reward functions and propose an $(1-\tfrac1e)$-approximation algorithm for the joint action selection by maximizing a sub-modular objective. MALinZero demonstrates state-of-the-art performance on multi-agent benchmarks such as matrix games, SMAC, and SMACv2, outperforming both model-based and model-free multi-agent reinforcement learning baselines with faster learning speed and better performance.

Multi-agent systems (MAS) built on Large Language Models (LLMs) are being used to approach complex problems and can surpass single model inference. However, their success hinges on navigating a fundamental cognitive tension: the need to balance broad, divergent exploration of the solution space with a principled, convergent synthesis to the optimal solution. Existing paradigms often struggle to manage this duality, leading to premature consensus, error propagation, and a critical credit assignment problem that fails to distinguish between genuine reasoning and superficially plausible arguments. To operationalize this critical synthesis phase, we introduce Conditional Listwise Policy Optimization (CLPO), a reinforcement learning objective that disentangles signals for strategic decisions and tactical rationales. By combining decision-focused policy gradients with a list-wise ranking loss over justifications, CLPO achieves clean credit assignment and stronger comparative supervision. The rise of large language models (LLMs) have enabled a new type of multi-agent system (MAS) (Park et al., 2023; Chen et al., 2023a; Zhu et al., 2025), where multiple model instances collaborate to tackle problems that exceed the capacity of any single model (Zhang et al., 2024a; Qiao et al., 2024; Han et al., 2025). By distributing roles and enabling structured interaction, MASs hold the promise of achieving robustness, creativity, and reliability that emerge from collective intelligence (Cheng et al., 2024; Pezeshkpour et al., 2024). At the heart of any effective collaborative system lies a fundamental cognitive tension. Early work in the psychology of creativity (Runco & Chand, 1995; Brophy, 2001; Zhang et al., 2020) emphasizes that intelligent problem-solving requires a dynamic balance between two seemingly contradictory modes of thought: Divergent Creativity and Convergent Critique. Guilford's theory of divergent and convergent thinking (Guilford, 1967) formalizes this duality: divergence is the generative process of exploring a wide array of alternative hypotheses, while convergence is the evaluative process of comparing, refining, and synthesizing these options.