Agentic AI systems, which leverage multiple autonomous agents and Large Language Models (LLMs), are increasingly used to address complex, multi-step tasks. The safety, security, and functionality of these systems are critical, especially in high-stakes applications. However, the current ecosystem of inter-agent communication is fragmented, with protocols such as the Model Context Protocol (MCP) for tool access and the Agent-to-Agent (A2A) protocol for coordination being analyzed in isolation. This fragmentation creates a semantic gap that prevents the rigorous analysis of system properties and introduces risks such as architectural misalignment and exploitable coordination issues. To address these challenges, we introduce a modeling framework for agentic AI systems composed of two foundational models. The first, the host agent model, formalizes the top-level entity that interacts with the user, decomposes tasks, and orchestrates their execution by leveraging external agents and tools. The second, the task lifecycle model, details the states and transitions of individual sub-tasks from creation to completion, providing a fine-grained view of task management and error handling. Together, these models provide a unified semantic framework for reasoning about the behavior of multi-AI agent systems. Grounded in this framework, we define 17 properties for the host agent and 14 for the task lifecycle, categorized into liveness, safety, completeness, and fairness. Expressed in temporal logic, these properties enable formal verification of system behavior, detection of coordination edge cases, and prevention of deadlocks and security vulnerabilities. Through this effort, we introduce the first rigorously grounded, domain-agnostic framework for the systematic analysis, design, and deployment of correct, reliable, and robust agentic AI systems.

Distributed denial-of-service (DDoS) attacks remain a critical threat to Internet services, causing costly disruptions. While machine learning (ML) has shown promise in DDoS detection, current solutions struggle with multi-domain environments where attacks must be detected across heterogeneous networks and organizational boundaries. This limitation severely impacts the practical deployment of ML-based defenses in real-world settings. This paper introduces Anomaly-Flow, a novel framework that addresses this critical gap by combining Federated Learning (FL) with Generative Adversarial Networks (GANs) for privacy-preserving, multi-domain DDoS detection. Our proposal enables collaborative learning across diverse network domains while preserving data privacy through synthetic flow generation. Through extensive evaluation across three distinct network datasets, Anomaly-Flow achieves an average F1-score of $0.747$, outperforming baseline models. Importantly, our framework enables organizations to share attack detection capabilities without exposing sensitive network data, making it particularly valuable for critical infrastructure and privacy-sensitive sectors. Beyond immediate technical contributions, this work provides insights into the challenges and opportunities in multi-domain DDoS detection, establishing a foundation for future research in collaborative network defense systems. Our findings have important implications for academic research and industry practitioners working to deploy practical ML-based security solutions.

Vertical Federated learning (VFL) is a class of FL where each client shares the same sample space but only holds a subset of the features. While VFL tackles key privacy challenges of distributed learning, it often assumes perfect hardware and communication capabilities. This assumption hinders the broad deployment of VFL, particularly on edge devices, which are heterogeneous in their in-situ capabilities and will connect/disconnect from the network over time. To address this gap, we define Internet Learning (IL) including its data splitting and network context and which puts good performance under extreme dynamic condition of clients as the primary goal. We propose VFL as a naive baseline and develop several extensions to handle the IL paradigm of learning. Furthermore, we implement new methods, propose metrics, and extensively analyze results based on simulating a sensor network. The results show that the developed methods are more robust to changes in the network than VFL baseline.

The current state-of-the-art large language models (LLMs) are effective in generating high-quality text and encapsulating a broad spectrum of world knowledge. However, these models often hallucinate during generation and are not designed to utilize external information sources. To enable requests to the external knowledge bases, also called knowledge grounding, retrieval-augmented LLMs were introduced. For now, their applications have largely involved Open Domain Question Answering, Abstractive Question Answering, and such. In this paper, we broaden the scope of retrieval-augmented LLMs by venturing into a new task - code generation using external entities. For this task, we collect and publish a new dataset for project-level code generation, where the model should reuse functions defined in the project during generation. As we show, existing retrieval-augmented LLMs fail to assign relevance scores between similar entity names, and to mitigate it, they expand entity names with description context and append it to the input. In practice, due to the limited context size they can not accommodate the indefinitely large context of the whole project. To solve this issue, we propose a novel end-to-end trainable architecture with an scalable entity retriever injected directly into the LLM decoder. We demonstrate that our model can outperform common baselines in several scenarios, including project-level code generation, as well as Bash and SQL scripting.

We propose a scheme based on active learning to reconstruct private strategies executed by a population of interacting agents and predict an exact outcome of the underlying multi-agent interaction process, here identified as a stationary action profile. We envision a scenario where an external observer, endowed with a learning procedure, can make queries and observe the agents' reactions through private action-reaction mappings, whose collective fixed point corresponds to a stationary profile. By iteratively collecting sensible data and updating parametric estimates of the action-reaction mappings, we establish sufficient conditions to assess the asymptotic properties of the proposed active learning methodology so that, if convergence happens, it can only be towards a stationary action profile. This fact yields two main consequences: i) learning locally-exact surrogates of the action-reaction mappings allows the external observer to succeed in its prediction task, and ii) working with assumptions so general that a stationary profile is not even guaranteed to exist, the established sufficient conditions hence act also as certificates for the existence of such a desirable profile. Extensive numerical simulations involving typical competitive multi-agent control and decision-making problems illustrate the practical effectiveness of the proposed learning-based approach. The authors are with the IMT School for Advanced Studies Lucca, Piazza San Francesco 19, 55100, Lucca, Italy ({filippo.fabiani,

Lack of trust: One of the biggest difficulty for AI or ML products is lack of trust. Millions of dollars have been spent on prototyping but with very little success in the real-world launches. Essentially, one of the most fundamental values of doing business and providing value to customers is trust, and Artificial Intelligence is the most-heavily debated technology when it comes to ethical concerns and related trust issues. Trust comes from involving different options and parties in the entire development phase, which is not done in the prototype phase. The complexity of a launch: Building a prototype is easy, but there are tens of other external entities that need to be considered when moving into the real world.

In Multi-Agent Systems (MAS) there are two main models of interaction: among agents, and between agents and the environment. Although there are studies considering these models, there is no practical tool to afford the interaction with external entities with both models. This paper presents a proposal for such a tool based on the Apache Camel framework by designing two new components, namely camel-jason and camel-artifact. By means of these components, an external entity is modelled according to its nature, i.e., whether it is autonomous or non-autonomous, interacting with the MAS respectively as an agent or an artifact. It models coherently external entities whereas Camel provides interoperability with several communication protocols.

The experience interacting with a Chatbot at many occasions for many of us probably was not that convincing despite it is being marketed with catchy tags like powered by AI, Smart & Intelligent, Self-Learning, Self-Enriching etc. And many people have already drawn their conclusions out of frustrations that Chatbots are not at all intelligent; and behind the scene a bunch of hard coded logics are working to facilitate the conversation- Which was probably partially true for some cases few years back. At large the story today is a bit different and we need to understand both the sides of the story to draw a fair conclusion. We will try to discuss different aspects of it one by one. But despite all these, Chatbots are still "Hot" and their adoption across all industries has increased many folds in past couple of years.

With the recent news about Facebook and Cambridge analytica, we are rightly concerned about the power and impact of algorithms to shape political debate and more generally, our lives. The social score model in China shows another way in which AI could influence all aspects of society. Based on these and other views, most policy makers in the West take a negative view of AI and the power of algorithms in society. In this post, I present a different, more optimistic view of the impact of AI on society where AI could be a part of the solution to overcome the problem of Algorithmocracy and filter bubbles. I discussed some of the ideas below Last week, I spoke at the Economist innovation summit in London.

Av3ar is the next generation Cognitive Computing and Machine Learning system from Perpetuuiti. Av3ar is the next generation Cognitive Computing and Machine Learning system from Perpetuuiti. It is aimed to deliver end-to-end Interactive solutions that dramatically improves the operational efficiencies of customers in the global marketplace. It understands, learns and responds back to customers with emotions like humans. The product gets plugged into the customer place and absorbs all the data to learn the pattern of the day-to-day tasks and processes.