The rapid advancement of Large Language Models (LLMs) has led to the emergence of Multi-Agent Systems (MAS) to perform complex tasks through collaboration. However, the intricate nature of MAS, including their architecture and agent interactions, raises significant concerns regarding intellectual property (IP) protection. In this paper, we introduce MASLEAK, a novel attack framework designed to extract sensitive information from MAS applications. MASLEAK targets a practical, black-box setting, where the adversary has no prior knowledge of the MAS architecture or agent configurations. The adversary can only interact with the MAS through its public API, submitting attack query $q$ and observing outputs from the final agent. Inspired by how computer worms propagate and infect vulnerable network hosts, MASLEAK carefully crafts adversarial query $q$ to elicit, propagate, and retain responses from each MAS agent that reveal a full set of proprietary components, including the number of agents, system topology, system prompts, task instructions, and tool usages. We construct the first synthetic dataset of MAS applications with 810 applications and also evaluate MASLEAK against real-world MAS applications, including Coze and CrewAI. MASLEAK achieves high accuracy in extracting MAS IP, with an average attack success rate of 87% for system prompts and task instructions, and 92% for system architecture in most cases. We conclude by discussing the implications of our findings and the potential defenses.

Influencer marketing has become a crucial feature of digital marketing strategies. Despite its rapid growth and algorithmic relevance, the field of computational studies in influencer marketing remains fragmented, especially with limited systematic reviews covering the computational methodologies employed. This makes overarching scientific measurements in the influencer economy very scarce, to the detriment of interested stakeholders outside of platforms themselves, such as regulators, but also researchers from other fields. This paper aims to provide an overview of the state of the art of computational studies in influencer marketing by conducting a systematic literature review (SLR) based on the PRISMA model. The paper analyses 69 studies to identify key research themes, methodologies, and future directions in this research field. The review identifies four major research themes: Influencer identification and characterisation, Advertising strategies and engagement, Sponsored content analysis and discovery, and Fairness. Methodologically, the studies are categorised into machine learning-based techniques (e.g., classification, clustering) and non-machine-learning-based techniques (e.g., statistical analysis, network analysis). Key findings reveal a strong focus on optimising commercial outcomes, with limited attention to regulatory compliance and ethical considerations. The review highlights the need for more nuanced computational research that incorporates contextual factors such as language, platform, and industry type, as well as improved model explainability and dataset reproducibility. The paper concludes by proposing a multidisciplinary research agenda that emphasises the need for further links to regulation and compliance technology, finer granularity in analysis, and the development of standardised datasets.

Despite impressive progress on complex reasoning, current large language models (LLMs) typically operate in isolation - treating each problem as an independent attempt, without accumulating or integrating experiential knowledge. In contrast, expert problem solvers - such as Olympiad or programming contest teams - leverage a rich tapestry of experiences: absorbing mentorship from coaches, developing intuition from past problems, leveraging knowledge of tool usage and library functionality, adapting strategies based on the expertise and experiences of peers, continuously refining their reasoning through trial and error, and learning from other related problems even during competition. We introduce Xolver, a training-free multi-agent reasoning framework that equips a black-box LLM with a persistent, evolving memory of holistic experience. Xolver integrates diverse experience modalities, including external and self-retrieval, tool use, collaborative interactions, agent-driven evaluation, and iterative refinement. By learning from relevant strategies, code fragments, and abstract reasoning patterns at inference time, Xolver avoids generating solutions from scratch - marking a transition from isolated inference toward experience-aware language agents. Built on both open-weight and proprietary models, Xolver consistently outperforms specialized reasoning agents. Even with lightweight backbones (e.g., QWQ-32B), it often surpasses advanced models including Qwen3-235B, Gemini 2.5 Pro, o3, and o4-mini-high. With o3-mini-high, it achieves new best results on GSM8K (98.1%), AIME'24 (94.4%), AIME'25 (93.7%), Math-500 (99.8%), and LiveCodeBench-V5 (91.6%) - highlighting holistic experience learning as a key step toward generalist agents capable of expert-level reasoning. Code and data are available at https://kagnlp.github.io/xolver.github.io/.

Optimizing urban freight logistics is critical for developing sustainable, low-carbon cities. Traditional methods often rely on manual coordination of simulation tools, optimization solvers, and expert-driven workflows, limiting their efficiency and scalability. This paper presents an agentic system architecture that leverages the model context protocol (MCP) to orchestrate multi-agent collaboration among scientific tools for autonomous, simulation-informed optimization in urban logistics. The system integrates generative AI agents with domain-specific engines - such as Gurobi for optimization and AnyLogic for agent-based simulation - forming a generative digital twin capable of reasoning, planning, and acting across multimodal freight networks. By incorporating integrated chatbots, retrieval-augmented generation, and structured memory, the framework enables agents to interpret user intent from natural language conversations, retrieve relevant datasets and models, coordinate solvers and simulators, and execute complex workflows. We demonstrate this approach through a freight decarbonization case study, showcasing how MCP enables modular, interoperable, and adaptive agent behavior across diverse toolchains. The results reveal that our system transforms digital twins from static visualizations into autonomous, decision-capable systems, advancing the frontiers of urban operations research. By enabling context-aware, generative agents to operate scientific tools automatically and collaboratively, this framework supports more intelligent, accessible, and dynamic decision-making in transportation planning and smart city management.

--This paper addresses the challenge of transforming complex sentences into sequences of logical, simplified sentences while preserving semantic and logical integrity with the help of Large Language Models. We propose a hybrid approach that combines advanced prompting with multi-agent architectures to enhance the sentence simplification process. Experimental results show that our approach was able to successfully simplify 70% of the complex sentences written for video game design application. In comparison, a single-agent approach attained a 48% success rate on the same task. Sentence simplification is a challenging task in computational linguistics. The simplification process aims to transform complex sentences into simpler structures while preserving the original meaning. Effective sentence simplification has significant applications across numerous domains like education, content accessibility for individuals with cognitive disabilities, automated content creation, robotics, coding, legal documents, etc. Traditional approaches to sentence simplification have relied on rule-based systems, statistical methods, and more recently neural network architectures [1]. Complex sentences present significant challenges in action-oriented contexts, particularly when attempting to derive executable/actionable functionalities such as robotics, legal documents, and video games.

Agentic tasks, which require multi-step problem solving with autonomy, tool use, and adaptive reasoning, are becoming increasingly central to the advancement of NLP and AI. However, existing instruction data lacks tool interaction, and current agentic benchmarks rely on costly human annotation, limiting their scalability. We introduce \textsc{TaskCraft}, an automated workflow for generating difficulty-scalable, multi-tool, and verifiable agentic tasks with execution trajectories. TaskCraft expands atomic tasks using depth-based and width-based extensions to create structurally and hierarchically complex challenges. Empirical results show that these tasks improve prompt optimization in the generation workflow and enhance supervised fine-tuning of agentic foundation models. We present a large-scale synthetic dataset of approximately 36,000 tasks with varying difficulty to support future research on agent tuning and evaluation.

Abstract--During collaborative tasks, human behavior is guided by multiple levels of intentions that evolve over time, such as task sequence preferences and interaction strategies. T o adapt to these changing preferences and promptly correct any inaccurate estimations, collaborative robots must accurately track these dynamic human intentions in real time. We propose a Hierarchical Intention Tracking (HIT) algorithm for collaborative robots to track dynamic and hierarchical human intentions effectively in real time. HIT represents human intentions as intention trees with arbitrary depth, and probabilistically tracks human intentions by Bayesian filtering, upward measurement propagation, and downward posterior propagation across all levels. We develop a HIT-based robotic system that dynamically switches between Interaction-Task and V erification-Task trees for a collaborative assembly task, allowing the robot to effectively coordinate human intentions at three levels: task-level (subtask goal locations), interaction-level (mode of engagement with the robot), and verification-level (confirming or correcting intention recognition). Our user study shows that our HIT-based collaborative robot system surpasses existing collaborative robot solutions by achieving a balance between efficiency, physical workload, and user comfort while ensuring safety and task completion. Post-experiment surveys further reveal that the HIT-based system enhances the user trust and minimizes interruptions to user's task flow through its effective understanding of human intentions across multiple levels. The video demonstrating our experiments is available at https://youtu.be/Y5kg7QC41yw. I. Introduction Robots require an effective understanding of human intentions to collaborate both safely and efficiently with humans. During long-term tasks, human intentions continuously evolve along with task progress. When handling a complex task, humans typically break down the task into milestones and sub-tasks at varying levels of granularity, leading to a hierarchical structure of human intentions. During collaboration, humans often maintain multiple intentions with different semantics simultaneously. For instance, they may prefer specific subtask sequences or modes of interaction with the robot (e.g.

One of the common artificial intelligence applications in electronic games consists of making an artificial agent learn how to execute some determined task successfully in a game environment. One way to perform this task is through machine learning algorithms capable of learning the sequence of actions required to win in a given game environment. There are several supervised learning techniques able to learn the correct answer for a problem through examples. However, when learning how to play electronic games, the correct answer might only be known by the end of the game, after all the actions were already taken. Thus, not being possible to measure the accuracy of each individual action to be taken at each time step. A way for dealing with this problem is through Neuroevolution, a method which trains Artificial Neural Networks using evolutionary algorithms. In this article, we introduce a framework for testing optimization algorithms with artificial agent controllers in electronic games, called EvoMan, which is inspired in the action-platformer game Mega Man II. The environment can be configured to run in different experiment modes, as single evolution, coevolution and others. To demonstrate some challenges regarding the proposed platform, as initial experiments we applied Neuroevolution using Genetic Algorithms and the NEAT algorithm, in the context of competitively coevolving two distinct agents in this game.

While knowledge distillation has become a mature field for compressing large language models (LLMs) into smaller ones by aligning their outputs or internal representations, the distillation of LLM-based agents, which involve planning, memory, and tool use, remains relatively underexplored. Existing agent distillation methods typically replay full teacher trajectories or imitate step-by-step teacher tool usage, but they often struggle to train student agents to dynamically plan and act in novel environments. We propose AgentDistill, a novel, training-free agent distillation framework that enables efficient and scalable knowledge transfer via direct reuse of Model-Context-Protocols (MCPs), which are structured and reusable task-solving modules autonomously generated by teacher agents. The reuse of these distilled MCPs enables student agents to generalize their capabilities across domains and solve new problems with minimal supervision or human intervention. Experiments on biomedical and mathematical benchmarks demonstrate that our distilled student agents, built on small language models, can achieve performance comparable to advanced systems using large LLMs such as OctoTools (GPT-4o), highlighting the effectiveness of our framework in building scalable and cost-efficient intelligent agents.

Effective decision-making on networks often relies on learning from graph-structured data, where Graph Neural Networks (GNNs) play a central role, but they take efforts to configure and tune. In this demo, we propose LLMNet, showing how to design GNN automated through Large Language Models. Our system develops a set of agents that construct graph-related knowlege bases and then leverages Retrieval-Augmented Generation (RAG) to support automated configuration and refinement of GNN models through a knowledge-guided evolution process. These agents, equipped with specialized knowledge bases, extract insights into tasks and graph structures by interacting with the knowledge bases. Empirical results show LLMNet excels in twelve datasets across three graph learning tasks, validating its effectiveness of GNN model designing.