This monograph presents a modular cognitive architecture for artificial intelligence grounded in the formal modeling of belief as structured semantic state. Belief states are defined as dynamic ensembles of linguistic expressions embedded within a navigable manifold, where operators enable assimilation, abstraction, nullification, memory, and introspection. Drawing from philosophy, cognitive science, and neuroscience, we develop a layered framework that enables self-regulating epistemic agents capable of reflective, goal-directed thought. At the core of this framework is the epistemic vacuum: a class of semantically inert cognitive states that serves as the conceptual origin of belief space. From this foundation, the Null Tower arises as a generative structure recursively built through internal representational capacities. The theoretical constructs are designed to be implementable in both symbolic and neural systems, including large language models, hybrid agents, and adaptive memory architectures. This work offers a foundational substrate for constructing agents that reason, remember, and regulate their beliefs in structured, interpretable ways.

Anthropic's first developer conference kicked off in San Francisco on Thursday, and while the rest of the industry races toward artificial general intelligence, at Anthropic the goal of the year is deploying a "virtual collaborator" in the form of an autonomous AI agent. "We're all going to have to contend with the idea that everything you do is eventually going to be done by AI systems," Anthropic CEO Dario Amodei said in a press briefing. As roughly 500 attendees munched breakfast sandwiches with an abnormal amount of arugula, and Anthropic staffers milled about in company-issued baseball caps, Amodei took the stage with his chief product officer, Mike Krieger. "When do you think there will be the first billion-dollar company with one human employee?" Amodei, wearing a light-gray jacket and a pair of Brooks running shoes, replied without skipping a beat: "2026."

LLM-based multi-agent systems (MAS) extend the capabilities of single LLMs by enabling cooperation among multiple specialized agents. However, most existing MAS frameworks rely on a single LLM to drive all agents, constraining the system's intelligence to the limit of that model. This paper explores the paradigm of heterogeneous LLM-driven MAS (X-MAS), where agents are powered by diverse LLMs, elevating the system's potential to the collective intelligence of diverse LLMs. We introduce X-MAS-Bench, a comprehensive testbed designed to evaluate the performance of various LLMs across different domains and MAS-related functions. As an extensive empirical study, we assess 27 LLMs across 5 domains (encompassing 21 test sets) and 5 functions, conducting over 1.7 million evaluations to identify optimal model selections for each domain-function combination. Building on these findings, we demonstrate that transitioning from homogeneous to heterogeneous LLM-driven MAS can significantly enhance system performance without requiring structural redesign. Specifically, in a chatbot-only MAS scenario, the heterogeneous configuration yields up to 8.4\% performance improvement on the MATH dataset. In a mixed chatbot-reasoner scenario, the heterogeneous MAS could achieve a remarkable 47\% performance boost on the AIME dataset. Our results underscore the transformative potential of heterogeneous LLMs in MAS, highlighting a promising avenue for advancing scalable, collaborative AI systems.

LLM-based multi-agent systems (MAS) have demonstrated significant potential in enhancing single LLMs to address complex and diverse tasks in practical applications. Despite considerable advancements, the field lacks a unified codebase that consolidates existing methods, resulting in redundant re-implementation efforts, unfair comparisons, and high entry barriers for researchers. To address these challenges, we introduce MASLab, a unified, comprehensive, and research-friendly codebase for LLM-based MAS. (1) MASLab integrates over 20 established methods across multiple domains, each rigorously validated by comparing step-by-step outputs with its official implementation. (2) MASLab provides a unified environment with various benchmarks for fair comparisons among methods, ensuring consistent inputs and standardized evaluation protocols. (3) MASLab implements methods within a shared streamlined structure, lowering the barriers for understanding and extension. Building on MASLab, we conduct extensive experiments covering 10+ benchmarks and 8 models, offering researchers a clear and comprehensive view of the current landscape of MAS methods. MASLab will continue to evolve, tracking the latest developments in the field, and invite contributions from the broader open-source community.

Single-agent LLMs hit hard limits--finite context, role overload, and brittle domain transfer. Conventional multi-agent fixes soften those edges yet expose fresh pains: ill-posed decompositions, fuzzy contracts, and verification overhead that blunts the gains. We therefore present Know-The-Ropes (KtR), a framework that converts domain priors into an algorithmic blueprint hierarchy, in which tasks are recursively split into typed, controller-mediated subtasks, each solved zero-shot or with the lightest viable boost (e.g., chain-of-thought, micro-tune, self-check). Grounded in the No-Free-Lunch theorem, KtR trades the chase for a universal prompt for disciplined decomposition. On the Knapsack problem (3-8 items), three GPT-4o-mini agents raise accuracy from 3% zero-shot to 95% on size-5 instances after patching a single bottleneck agent. On the tougher Task-Assignment problem (6-15 jobs), a six-agent o3-mini blueprint hits 100% up to size 10 and 84% on sizes 13-15, versus 11% zero-shot. Algorithm-aware decomposition plus targeted augmentation thus turns modest models into reliable collaborators--no ever-larger monoliths required.

--The growing complexity of power systems has made accurate load forecasting more important than ever . An increasing number of advanced load forecasting methods have been developed. However, the static design of current methods offers no mechanism for human-model interaction. As the primary users of forecasting models, system operators often find it difficult to understand and apply these advanced models, which typically requires expertise in artificial intelligence (AI). This also prevents them from incorporating their experience and real-world contextual understanding into the forecasting process. Recent breakthroughs in large language models (LLMs) offer a new opportunity to address this issue. By leveraging their natural language understanding and reasoning capabilities, we propose an LLM-based multi-agent collaboration framework to bridge the gap between human operators and forecasting models. A set of specialized agents is designed to perform different tasks in the forecasting workflow and collaborate via a dedicated communication mechanism. Our experiments demonstrate that the interactive load forecasting accuracy can be significantly improved when users provide proper insight in key stages. Our cost analysis shows that the framework remains affordable, making it practical for real-world deployment. With the boom of artificial intelligence, a wide range of forecasting algorithms have been proposed recently, many of which have demonstrated impressive performance. However, these forecasting methods become static once designed, offering no mechanism for interaction between the model and human users. This lack of interaction creates major barriers to the practical use of the forecasting methods.

The serious games between humans and AI have only just begun. Evolutionary Game Theory (EGT) models the competitive and cooperative strategies of biological entities. EGT could help predict the potential evolutionary equilibrium of humans and AI. The objective of this work was to examine some of the EGT models relevant to human-AI interaction, evolution, and coevolution. Of thirteen EGT models considered, three were examined: the Hawk-Dove Game, Iterated Prisoner's Dilemma, and the War of Attrition. This selection was based on the widespread acceptance and clear relevance of these models to potential human-AI evolutionary dynamics and coevolutionary trajectories. The Hawk-Dove Game predicts balanced mixed-strategy equilibria based on the costs of conflict. It also shows the potential for balanced coevolution rather than dominance. Iterated Prisoner's Dilemma suggests that repeated interaction may lead to cognitive coevolution. It demonstrates how memory and reciprocity can lead to cooperation. The War of Attrition suggests that competition for resources may result in strategic coevolution, asymmetric equilibria, and conventions on sharing resources. Therefore, EGT may provide a suitable framework to understand and predict the human-AI evolutionary dynamic. However, future research could extend beyond EGT and explore additional frameworks, empirical validation methods, and interdisciplinary perspectives. AI is being shaped by human input and is evolving in response to it. So too, neuroplasticity allows the human brain to grow and evolve in response to stimuli. If humans and AI converge in future, what might be the result of human neuroplasticity combined with an ever-evolving AI? Future research should be mindful of the ethical and cognitive implications of human-AI interaction, evolution, and coevolution.

In many real-world complex systems, the behavior can be observed as a collection of discrete events generated by multiple interacting agents. Analyzing the dynamics of these multi-agent systems, especially team sports, often relies on understanding the movement and interactions of individual agents. However, while providing valuable snapshots, event-based positional data typically lacks the continuous temporal information needed to directly calculate crucial properties such as velocity. This absence severely limits the depth of dynamic analysis, preventing a comprehensive understanding of individual agent behaviors and emergent team strategies. To address this challenge, we propose a new method to simultaneously complete the velocity of all agents using only the event-based positional data from team sports. Based on this completed velocity information, we investigate the applicability of existing team sports analysis and evaluation methods. Experiments using soccer event data demonstrate that neural network-based approaches outperformed rule-based methods regarding velocity completion error, considering the underlying temporal dependencies and graph structure of player-to-player or player-to-ball interaction. Moreover, the space evaluation results obtained using the completed velocity are closer to those derived from complete tracking data, highlighting our method's potential for enhanced team sports system analysis.

The emergence of Large Language Models (LLMs) has reshaped agent systems. Unlike traditional rule-based agents with limited task scope, LLM-powered agents offer greater flexibility, cross-domain reasoning, and natural language interaction. Moreover, with the integration of multi-modal LLMs, current agent systems are highly capable of processing diverse data modalities, including text, images, audio, and structured tabular data, enabling richer and more adaptive real-world behavior. This paper comprehensively examines the evolution of agent systems from the pre-LLM era to current LLM-powered architectures. We categorize agent systems into software-based, physical, and adaptive hybrid systems, highlighting applications across customer service, software development, manufacturing automation, personalized education, financial trading, and healthcare. We further discuss the primary challenges posed by LLM-powered agents, including high inference latency, output uncertainty, lack of evaluation metrics, and security vulnerabilities, and propose potential solutions to mitigate these concerns.

Validating scientific hypotheses is a central challenge in biomedical research, and remains difficult for artificial intelligence (AI) agents due to the complexity of real-world data analysis and evidence interpretation. In this work, we present BioDSA-1K, a benchmark designed to evaluate AI agents on realistic, data-driven biomedical hypothesis validation tasks. BioDSA-1K consists of 1,029 hypothesis-centric tasks paired with 1,177 analysis plans, curated from over 300 published biomedical studies to reflect the structure and reasoning found in authentic research workflows. Each task includes a structured hypothesis derived from the original study's conclusions, expressed in the affirmative to reflect the language of scientific reporting, and one or more pieces of supporting evidence grounded in empirical data tables. While these hypotheses mirror published claims, they remain testable using standard statistical or machine learning methods. The benchmark enables evaluation along four axes: (1) hypothesis decision accuracy, (2) alignment between evidence and conclusion, (3) correctness of the reasoning process, and (4) executability of the AI-generated analysis code. Importantly, BioDSA-1K includes non-verifiable hypotheses: cases where the available data are insufficient to support or refute a claim, reflecting a common yet underexplored scenario in real-world science. We propose BioDSA-1K as a foundation for building and evaluating generalizable, trustworthy AI agents for biomedical discovery.