The field of AI is undergoing a fundamental transition from systems that can produce synthetic content upon request to autonomous agents that can plan and execute complex tasks with only limited human involvement. Companies that pioneered the development of generative AI tools are now building AI agents that can be instructed to independently navigate the internet, perform a wide range of online tasks, and serve as artificial personal assistants and virtual coworkers. The opportunities presented by this new technology are tremendous, as are the associated risks. Fortunately, there exist robust analytic frameworks for confronting many of these challenges, namely, the economic theory of principal-agent problems and the common law doctrine of agency relationships. Drawing on these frameworks, this Article makes three contributions. First, it uses agency law and theory to identify and characterize problems arising from AI agents, including issues of information asymmetry, discretionary authority, and loyalty. Second, it illustrates the limitations of conventional solutions to agency problems: incentive design, monitoring, and enforcement might not be effective for governing AI agents that make uninterpretable decisions and operate at unprecedented speed and scale. Third, the Article explores the implications of agency law and theory for designing and regulating AI agents, arguing that new technical and legal infrastructure is needed to support governance principles of inclusivity, visibility, and liability.

Traditional interactive environments limit agents' intelligence growth with fixed tasks. Recently, single-agent environments address this by generating new tasks based on agent actions, enhancing task diversity. We consider the decision-making problem in multi-agent settings, where tasks are further influenced by social connections, affecting rewards and information access. However, existing multi-agent environments lack a combination of adaptive physical surroundings and social connections, hindering the learning of intelligent behaviors. To address this, we introduce AdaSociety, a customizable multi-agent environment featuring expanding state and action spaces, alongside explicit and alterable social structures. As agents progress, the environment adaptively generates new tasks with social structures for agents to undertake. In AdaSociety, we develop three mini-games showcasing distinct social structures and tasks. Initial results demonstrate that specific social structures can promote both individual and collective benefits, though current reinforcement learning and LLM-based algorithms show limited effectiveness in leveraging social structures to enhance performance. Overall, AdaSociety serves as a valuable research platform for exploring intelligence in diverse physical and social settings.

Federated Learning (FL) has emerged as a powerful collaborative learning paradigm where multiple clients jointly train a machine learning model without sharing their local data. In the classical FL setting, a central server coordinates multiple clients (e.g., mobile devices, edge devices) to collaboratively learn a shared global model without exchanging their own training data [48, 54, 79, 64]. In this scenario, each client performs local computations on its private data and periodically communicates model updates to the server, which aggregates them to update the global model. This collaborative approach has been successfully applied in various domains, including natural language processing [69, 43], computer vision [70, 63], and healthcare [4, 116]. Despite their success, traditional FL frameworks rely on the key assumption that all participants are fully cooperative and share aligned objectives, collectively working towards optimizing the performance of a shared global model (e.g., minimizing the average of individual loss functions). This assumption overlooks situations where participants have individual objectives, or competitive interests that may not align with the collective goal. Diverse examples of such scenarios have been extensively considered in the game theory literature, including Cournot competition in economics [2], optical networks [91], electricity markets [98], energy consumption control in smart grid [120], or mobile robot control [49]. Despite their relevance, these applications have yet to be associated with FL, presenting an unexplored opportunity to bridge game theory and FL for more robust and realistic frameworks.

The field of artificial intelligence has witnessed significant advancements in natural language processing, largely attributed to the capabilities of Large Language Models (LLMs). These models form the backbone of Agents designed to address long-context dependencies, particularly in Document-level Machine Translation (DocMT). DocMT presents unique challenges, with quality, consistency, and fluency being the key metrics for evaluation. Existing approaches, such as Doc2Doc and Doc2Sent, either omit sentences or compromise fluency. This paper introduces Doc-Guided Sent2Sent++, an Agent that employs an incremental sentence-level forced decoding strategy \textbf{to ensure every sentence is translated while enhancing the fluency of adjacent sentences.} Our Agent leverages a Doc-Guided Memory, focusing solely on the summary and its translation, which we find to be an efficient approach to maintaining consistency. Through extensive testing across multiple languages and domains, we demonstrate that Sent2Sent++ outperforms other methods in terms of quality, consistency, and fluency. The results indicate that, our approach has achieved significant improvements in metrics such as s-COMET, d-COMET, LTCR-$1_f$, and document-level perplexity (d-ppl). The contributions of this paper include a detailed analysis of current DocMT research, the introduction of the Sent2Sent++ decoding method, the Doc-Guided Memory mechanism, and validation of its effectiveness across languages and domains.

A common assumption in the literature on information diffusion is that populations are homogeneous regarding individuals' information acquisition and propagation process: Individuals update their informed and actively communicating state either through imitation (simple contagion) or peer influence (complex contagion). Here, we study the impact of the mixing and placement of individuals with different update processes on how information cascades in social networks. We consider Simple Spreaders, which take information from a random neighbor and communicate it, and Threshold-based Spreaders, which require a threshold number of active neighbors to change their state to active communication. Even though, in a population made exclusively of Simple Spreaders, information reaches all elements of any (connected) network, we show that, when Simple and Threshold-based Spreaders coexist and occupy random positions in a social network, the number of Simple Spreaders systematically amplifies the cascades only in degree heterogeneous networks (exponential and scale-free). In random and modular structures, this cascading effect originated by Simple Spreaders only exists above a critical mass of these individuals. In contrast, when Threshold-based Spreaders are assorted preferentially in the nodes with a higher degree, the cascading effect of Simple Spreaders vanishes, and the spread of information is drastically impaired. Overall, the study highlights the significance of the strategic placement of different roles in networked structures, with Simple Spreaders driving widespread cascades in heterogeneous networks and Threshold-based Spreaders playing a critical regulatory role in information spread with a tunable effect based on the threshold value. These effects have consequences to our understanding of social phenomena, such as the spread of innovations in heterogeneous social systems with the presence of eager (Simple Spreaders) versus averse (Threshold-based Spreaders) adopters, but also to information warfare on social media where Simple Spreaders can be seen as embedded agents (e.g., bots) used to amplify the virality of ill-intended content and, oppositely, Threshold-based Spreaders as an essential self-regulatory element of social systems operating as information filters.

The Alzheimer's Disease Analysis Model Generation 1 (ADAM) is a multi-agent large language model (LLM) framework designed to integrate and analyze multi-modal data, including microbiome profiles, clinical datasets, and external knowledge bases, to enhance the understanding and detection of Alzheimer's disease (AD). By leveraging retrieval-augmented generation (RAG) techniques along with its multi-agent architecture, ADAM-1 synthesizes insights from diverse data sources and contextualizes findings using literature-driven evidence. Comparative evaluation against XGBoost revealed similar mean F1 scores but significantly reduced variance for ADAM-1, highlighting its robustness and consistency, particularly in small laboratory datasets. While currently tailored for binary classification tasks, future iterations aim to incorporate additional data modalities, such as neuroimaging and biomarkers, to broaden the scalability and applicability for Alzheimer's research and diagnostics.

Cloud Operations (CloudOps) is a rapidly growing field focused on the automated management and optimization of cloud infrastructure which is essential for organizations navigating increasingly complex cloud environments. MontyCloud Inc. is one of the major companies in the CloudOps domain that leverages autonomous bots to manage cloud compliance, security, and continuous operations. To make the platform more accessible and effective to the customers, we leveraged the use of GenAI. Developing a GenAI-based solution for autonomous CloudOps for the existing MontyCloud system presented us with various challenges such as i) diverse data sources; ii) orchestration of multiple processes; and iii) handling complex workflows to automate routine tasks. To this end, we developed MOYA, a multi-agent framework that leverages GenAI and balances autonomy with the necessary human control. This framework integrates various internal and external systems and is optimized for factors like task orchestration, security, and error mitigation while producing accurate, reliable, and relevant insights by utilizing Retrieval Augmented Generation (RAG). Evaluations of our multi-agent system with the help of practitioners as well as using automated checks demonstrate enhanced accuracy, responsiveness, and effectiveness over non-agentic approaches across complex workflows.

This paper addresses a critical challenge in the high-speed passenger railway industry: designing effective dynamic pricing strategies in the context of competing and cooperating operators. To address this, a multi-agent reinforcement learning (MARL) framework based on a non-zero-sum Markov game is proposed, incorporating random utility models to capture passenger decision making. Unlike prior studies in areas such as energy, airlines, and mobile networks, dynamic pricing for railway systems using deep reinforcement learning has received limited attention. A key contribution of this paper is a parametrisable and versatile reinforcement learning simulator designed to model a variety of railway network configurations and demand patterns while enabling realistic, microscopic modelling of user behaviour, called RailPricing-RL. This environment supports the proposed MARL framework, which models heterogeneous agents competing to maximise individual profits while fostering cooperative behaviour to synchronise connecting services. Experimental results validate the framework, demonstrating how user preferences affect MARL performance and how pricing policies influence passenger choices, utility, and overall system dynamics. This study provides a foundation for advancing dynamic pricing strategies in railway systems, aligning profitability with system-wide efficiency, and supporting future research on optimising pricing policies.

The effective design of patrol strategies is a difficult and complex problem, especially in medium and large areas. The objective is to plan, in a coordinated manner, the optimal routes for a set of patrols in a given area, in order to achieve maximum coverage of the area, while also trying to minimize the number of patrols. In this paper, we propose a multi-agent reinforcement learning (MARL) model, based on a decentralized partially observable Markov decision process, to plan unpredictable patrol routes within an urban environment represented as an undirected graph. The model attempts to maximize a target function that characterizes the environment within a given time frame. Our model has been tested to optimize police patrol routes in three medium-sized districts of the city of Malaga. The aim was to maximize surveillance coverage of the most crime-prone areas, based on actual crime data in the city. To address this problem, several MARL algorithms have been studied, and among these the Value Decomposition Proximal Policy Optimization (VDPPO) algorithm exhibited the best performance. We also introduce a novel metric, the coverage index, for the evaluation of the coverage performance of the routes generated by our model. This metric is inspired by the predictive accuracy index (PAI), which is commonly used in criminology to detect hotspots. Using this metric, we have evaluated the model under various scenarios in which the number of agents (or patrols), their starting positions, and the level of information they can observe in the environment have been modified. Results show that the coordinated routes generated by our model achieve a coverage of more than $90\%$ of the $3\%$ of graph nodes with the highest crime incidence, and $65\%$ for $20\%$ of these nodes; $3\%$ and $20\%$ represent the coverage standards for police resource allocation.

This paper provides the first formalization of self-interested planning in multiagent settings using expectation-maximization (EM). Our formalization in the context of infinite-horizon and finitely-nested interactive POMDPs (I-POMDP) is distinct from EM formulations for POMDPs and cooperative multiagent planning frameworks. We exploit the graphical model structure specific to I-POMDPs, and present a new approach based on block-coordinate descent for further speed up. Forward filtering-backward sampling -- a combination of exact filtering with sampling -- is explored to exploit problem structure.