Deep learning has advanced medical image classification, but interpretability challenges hinder its clinical adoption. This study enhances interpretability in Chest X-ray (CXR) classification by using concept bottleneck models (CBMs) and a multi-agent Retrieval-Augmented Generation (RAG) system for report generation. By modeling relationships between visual features and clinical concepts, we create interpretable concept vectors that guide a multi-agent RAG system to generate radiology reports, enhancing clinical relevance, explainability, and transparency. Evaluation of the generated reports using an LLM-as-a-judge confirmed the interpretability and clinical utility of our model's outputs. On the COVID-QU dataset, our model achieved 81% classification accuracy and demonstrated robust report generation performance, with five key metrics ranging between 84% and 90%. This interpretable multi-agent framework bridges the gap between high-performance AI and the explainability required for reliable AI-driven CXR analysis in clinical settings.

This paper introduces state abstraction for two-player zero-sum Markov games (TZMGs), where the payoffs for the two players are determined by the state representing the environment and their respective actions, with state transitions following Markov decision processes. For example, in games like soccer, the value of actions changes according to the state of play, and thus such games should be described as Markov games. In TZMGs, as the number of states increases, computing equilibria becomes more difficult. Therefore, we consider state abstraction, which reduces the number of states by treating multiple different states as a single state. There is a substantial body of research on finding optimal policies for Markov decision processes using state abstraction. However, in the multi-player setting, the game with state abstraction may yield different equilibrium solutions from those of the ground game. To evaluate the equilibrium solutions of the game with state abstraction, we derived bounds on the duality gap, which represents the distance from the equilibrium solutions of the ground game. Finally, we demonstrate our state abstraction with Markov Soccer, compute equilibrium policies, and examine the results.

Recent advancements in language models (LMs) have sparked growing interest in developing LM agents. While fully autonomous agents could excel in many scenarios, numerous use cases inherently require them to collaborate with humans due to humans' latent preferences, domain expertise, or need for control. To facilitate the study of human-agent collaboration, we present Collaborative Gym (Co-Gym), a general framework enabling asynchronous, tripartite interaction among agents, humans, and task environments. We instantiate Co-Gym with three representative tasks in both simulated and real-world conditions, and propose an evaluation framework that assesses both the collaboration outcomes and processes. Our findings reveal that collaborative agents consistently outperform their fully autonomous counterparts in task performance within those delivered cases, achieving win rates of 86% in Travel Planning, 74% in Tabular Analysis, and 66% in Related Work when evaluated by real users. However, our study also highlights significant challenges in developing collaborative agents, requiring advancements in core aspects of intelligence -- communication capabilities, situational awareness, and balancing autonomy and human control.

In multi-agent environments, agents often struggle to learn optimal policies due to sparse or delayed global rewards, particularly in long-horizon tasks where it is challenging to evaluate actions at intermediate time steps. We introduce Temporal-Agent Reward Redistribution (TAR$^2$), a novel approach designed to address the agent-temporal credit assignment problem by redistributing sparse rewards both temporally and across agents. TAR$^2$ decomposes sparse global rewards into time-step-specific rewards and calculates agent-specific contributions to these rewards. We theoretically prove that TAR$^2$ is equivalent to potential-based reward shaping, ensuring that the optimal policy remains unchanged. Empirical results demonstrate that TAR$^2$ stabilizes and accelerates the learning process. Additionally, we show that when TAR$^2$ is integrated with single-agent reinforcement learning algorithms, it performs as well as or better than traditional multi-agent reinforcement learning methods.

Left-behind children (LBCs), numbering over 66 million in China, face severe mental health challenges due to parental migration for work. Early screening and identification of at-risk LBCs is crucial, yet challenging due to the severe shortage of mental health professionals, especially in rural areas. While the House-Tree-Person (HTP) test shows higher child participation rates, its requirement for expert interpretation limits its application in resource-scarce regions. To address this challenge, we propose PsyDraw, a multi-agent system based on Multimodal Large Language Models that assists mental health professionals in analyzing HTP drawings. The system employs specialized agents for feature extraction and psychological interpretation, operating in two stages: comprehensive feature analysis and professional report generation. Evaluation of HTP drawings from 290 primary school students reveals that 71.03% of the analyzes achieved High Consistency with professional evaluations, 26.21% Moderate Consistency and only 2.41% Low Consistency. The system identified 31.03% of cases requiring professional attention, demonstrating its effectiveness as a preliminary screening tool. Currently deployed in pilot schools, \method shows promise in supporting mental health professionals, particularly in resource-limited areas, while maintaining high professional standards in psychological assessment.

Smart spaces are ubiquitous computing environments that integrate diverse sensing and communication technologies to enhance space functionality, optimize energy utilization, and improve user comfort and well-being. The integration of emerging AI methodologies into these environments facilitates the formation of AI-driven smart spaces, which further enhance functionalities of the spaces by enabling advanced applications such as personalized comfort settings, interactive living spaces, and automatization of the space systems, all resulting in enhanced indoor experiences of the users. In this paper, we present a systematic survey of existing research on the foundational components of AI-driven smart spaces, including sensor technologies, data communication protocols, sensor network management and maintenance strategies, as well as the data collection, processing and analytics. Given the pivotal role of AI in establishing AI-powered smart spaces, we explore the opportunities and challenges associated with traditional machine learning (ML) approaches, such as deep learning (DL), and emerging methodologies including large language models (LLMs). Finally, we provide key insights necessary for the development of AI-driven smart spaces, propose future research directions, and sheds light on the path forward.

Federated reinforcement learning (FRL) has emerged as a promising paradigm, enabling multiple agents to collaborate and learn a shared policy adaptable across heterogeneous environments. Among the various reinforcement learning (RL) algorithms, the actor-critic (AC) algorithm stands out for its low variance and high sample efficiency. However, little to nothing is known theoretically about AC in a federated manner, especially each agent interacts with a potentially different environment. The lack of such results is attributed to various technical challenges: a two-level structure illustrating the coupling effect between the actor and the critic, heterogeneous environments, Markovian sampling and multiple local updates. In response, we study \textit{Single-loop Federated Actor Critic} (SFAC) where agents perform actor-critic learning in a two-level federated manner while interacting with heterogeneous environments. We then provide bounds on the convergence error of SFAC. The results show that the convergence error asymptotically converges to a near-stationary point, with the extent proportional to environment heterogeneity. Moreover, the sample complexity exhibits a linear speed-up through the federation of agents. We evaluate the performance of SFAC through numerical experiments using common RL benchmarks, which demonstrate its effectiveness.

As the demand for artificial intelligence (AI) grows to address complex real-world tasks, single models are often insufficient, requiring the integration of multiple models into pipelines. This paper introduces Bel Esprit, a conversational agent designed to construct AI model pipelines based on user-defined requirements. Bel Esprit employs a multi-agent framework where subagents collaborate to clarify requirements, build, validate, and populate pipelines with appropriate models. We demonstrate the effectiveness of this framework in generating pipelines from ambiguous user queries, using both human-curated and synthetic data. A detailed error analysis highlights ongoing challenges in pipeline construction. Bel Esprit is available for a free trial at https://belesprit.aixplain.com.

We study the problem of realizing strategies for an LTLf goal specification while ensuring that at least an LTLf backup specification is satisfied in case of unreliability of certain input variables. We formally define the problem and characterize its worst-case complexity as 2EXPTIME-complete, like standard LTLf synthesis. Then we devise three different solution techniques: one based on direct automata manipulation, which is 2EXPTIME, one disregarding unreliable input variables by adopting a belief construction, which is 3EXPTIME, and one leveraging second-order quantified LTLf (QLTLf), which is 2EXPTIME and allows for a direct encoding into monadic second-order logic, which in turn is worst-case nonelementary. We prove their correctness and evaluate them against each other empirically. Interestingly, theoretical worst-case bounds do not translate into observed performance; the MSO technique performs best, followed by belief construction and direct automata manipulation. As a byproduct of our study, we provide a general synthesis procedure for arbitrary QLTLf specifications.

Inspection of infrastructure using static sensor nodes has become a well established approach in recent decades. In this work, we present an experimental setup to address a binary inspection task using mobile sensor nodes. The objective is to identify the predominant tile type in a 1mx1m tiled surface composed of vibrating and non-vibrating tiles. A swarm of miniaturized robots, equipped with onboard IMUs for sensing and IR sensors for collision avoidance, performs the inspection. The decision-making approach leverages a Bayesian algorithm, updating robots' belief using inference. The original algorithm uses one of two information sharing strategies. We introduce a novel information sharing strategy, aiming to accelerate the decision-making. To optimize the algorithm parameters, we develop a simulation framework calibrated to our real-world setup in the high-fidelity Webots robotic simulator. We evaluate the three information sharing strategies through simulations and real-world experiments. Moreover, we test the effectiveness of our optimization by placing swarms with optimized and non-optimized parameters in increasingly complex environments with varied spatial correlation and fill ratios. Results show that our proposed information sharing strategy consistently outperforms previously established information-sharing strategies in decision time. Additionally, optimized parameters yield robust performance across different environments. Conversely, non-optimized parameters perform well in simpler scenarios but show reduced accuracy in complex settings.