Recent advances in language model (LM) agents have demonstrated significant potential for automating complex real-world tasks. To make progress on these difficult tasks, LM agent architectures have become increasingly complex, often incorporating multi-step retrieval tools, multiple agents, and scaffolding adapted to the underlying LM. In this work, we investigate whether all of this complexity is necessary, or if parts of these scaffolds can be removed on challenging tasks like SWE-bench. We show that in the case of SWE-bench, simply putting the entire environment into the context of a long context language model (LCLM) and properly prompting the model makes it competitive with carefully tuned, complex agent scaffolds. We show that a Gemini-1.5-Pro model without any scaffolding or tools achieves 38% on SWE-Bench-Verified, comparable with approaches using carefully tuned agent scaffolds (32%). While the unscaffolded approach with Gemini-1.5-Pro falls short of the strongest agentic architectures, we demonstrate that the more capable Gemini-2.5-Pro using the same unscaffolded approach directly attains a 50.8% solve rate. Additionally, a two-stage approach combining Gemini-1.5-Pro with Claude-3.7 achieves a competitive 48.6% solve rate.

Applying machine learning methods to physical systems that are supposed to act in the real world requires providing safety guarantees. However, methods that include such guarantees often come at a high computational cost, making them inapplicable to large datasets and embedded devices with low computational power. In this paper, we propose CoLSafe, a computationally lightweight safe learning algorithm whose computational complexity grows sublinearly with the number of data points. We derive both safety and optimality guarantees and showcase the effectiveness of our algorithm on a seven-degrees-of-freedom robot arm.

Microstructure often dictates materials performance, yet it is rarely treated as an explicit design variable because microstructure is hard to quantify, predict, and optimize . Here, w e introduce an image centric, closed - loop framework that makes microstructural morphology into a controllable objective and demonstrate its use case with Li - and Mn - rich layered oxide cathode precursors. This work present s an integrated, AI driven framework for the predictive design and optimization of lithium - ion battery cathode precursor synthesis. This framework integrates a diffusion - based image generation model, a quantitative image analysis pipeline, and a particle swarm optimization (PSO) algorithm. By extracting key morphological descriptors such as texture, s phericity, and median particle size (D) from SEM images, the platform accurately predicts SEM like morphologies resulting from specific coprecipitation conditions, including reaction time -, solution concentration -, and pH - dependent structural changes. Optimization then pinpoints synthesis parameters that yield user defined target morphologies, as experimentally validated by the close agreement between predicted and synthesized structures. This framework offers a practical strategy for data driven material s design, enabling both forward prediction and inverse design of synthesis conditions and paving the way toward autonomous, image guided microstructure engineering.

Software issue localization, the task of identifying the precise code locations (files, classes, or functions) relevant to a natural language issue description (e.g., bug report, feature request), is a critical yet time-consuming aspect of software development. While recent LLM-based agentic approaches demonstrate promise, they often incur significant latency and cost due to complex multi-step reasoning and relying on closed-source LLMs. Alternatively, traditional code ranking models, typically optimized for query-to-code or code-to-code retrieval, struggle with the verbose and failure-descriptive nature of issue localization queries. To bridge this gap, we introduce SweRank, an efficient and effective retrieve-and-rerank framework for software issue localization. To facilitate training, we construct SweLoc, a large-scale dataset curated from public GitHub repositories, featuring real-world issue descriptions paired with corresponding code modifications. Empirical results on SWE-Bench-Lite and LocBench show that SweRank achieves state-of-the-art performance, outperforming both prior ranking models and costly agent-based systems using closed-source LLMs like Claude-3.5. Further, we demonstrate SweLoc's utility in enhancing various existing retriever and reranker models for issue localization, establishing the dataset as a valuable resource for the community.

What makes a society possible at all? How is coordination and cooperation in social activity possible? What is the minimal mental architecture of a social agent? How is the information about the state of the world related to the agents intentions? How are the intentions of agents related? What role does communication play in this coordination process? This essay explores the conceptual and logical foundations of artificial social intelligence in the context of a society of multiple agents that communicate and cooperate to achieve some end. An attempt is made to provide an introduction to some of the key concepts, their formal definitions and their interrelationships. These include the notion of a changing social world of multiple agents. The logic of social intelligence goes beyond classical logic by linking information with strategic thought. A minimal architecture of social agents is presented. The agents have different dynamically changing, possible choices and abilities. The agents also have uncertainty, lacking perfect information about their physical state as well as their dynamic social state. The social state of an agent includes the intentional state of that agent, as well as, that agent's representation of the intentional states of other agents. Furthermore, it includes the evaluations agents make of their physical and social condition. Communication, semantic and pragmatic meaning and their relationship to intention and information states are investigated. The logic of agent abilities and intentions are motivated and formalized. The entropy of group strategic states is defined.

As Radio Access Networks (RAN) evolve toward AI-native architectures, intelligent modules such as xApps and rApps are expected to make increasingly autonomous decisions across scheduling, mobility, and resource management domains. However, these agents remain fundamentally stateless, treating each decision as isolated, lacking any persistent memory of prior events or outcomes. This reactive behavior constrains optimization, especially in environments where network dynamics exhibit episodic or recurring patterns. In this work, we propose RAN Cortex, a memory-augmented architecture that enables contextual recall in AI-based RAN decision systems. RAN Cortex introduces a modular layer composed of four elements: a context encoder that transforms network state into high-dimensional embeddings, a vector-based memory store of past network episodes, a recall engine to retrieve semantically similar situations, and a policy interface that supplies historical context to AI agents in real time or near-real time. We formalize the retrieval-augmented decision problem in the RAN, present a system architecture compatible with O-RAN interfaces, and analyze feasible deployments within the Non-RT and Near-RT RIC domains. Through illustrative use cases such as stadium traffic mitigation and mobility management in drone corridors, we demonstrate how contextual memory improves adaptability, continuity, and overall RAN intelligence. This work introduces memory as a missing primitive in AI-native RAN designs and provides a framework to enable "learning agents" without the need for retraining or centralized inference

The transition from monolithic to microservices architecture revolutionized software development by improving scalability and maintainability. This paradigm shift is now becoming relevant for complex multi-agent systems (MAS). This review article explores the evolution from monolithic architecture to microservices architecture in the specific context of MAS. It will highlight the limitations of traditional monolithic MAS and the benefits of adopting a microservices-based approach. The article further examines the core architectural principles and communication protocols, including Agent Communication Languages (ACLs), the Model Context Protocol (MCP), and the Application-to-Application (A2A) protocol. The article identifies emerging architectural patterns, design challenges, and considerations through a comparative lens of the paradigm shift.

This paper introduces MACH, a novel approach for optimizing task handover in vehicular computing scenarios. To ensure fast and latency-aware placement of tasks, the decision-making -- where and when should tasks be offloaded -- is carried out decentralized at the Road Side Units (RSUs) who also execute the tasks. By shifting control to the network edge, MACH moves away from the traditional centralized or vehicle-based handover method. Still, it focuses on contextual factors, such as the current RSU load and vehicle trajectories. Thus, MACH improves the overall Quality of Service (QoS) while fairly balancing computational loads between RSUs. To evaluate the effectiveness of our approach, we develop a robust simulation environment composed of real-world traffic data, dynamic network conditions, and different infrastructure capacities. For scenarios that demand low latency and high reliability, our experimental results demonstrate how MACH significantly improves the adaptability and efficiency of vehicular computations. By decentralizing control to the network edge, MACH effectively reduces communication overhead and optimizes resource utilization, offering a robust framework for task handover management.

Microgrids have emerged as a pivotal solution in the quest for a sustainable and energy-efficient future. While microgrids offer numerous advantages, they are also prone to issues related to reliably forecasting renewable energy demand and production, protecting against cyberattacks, controlling operational costs, optimizing power flow, and regulating the performance of energy management systems (EMS). Tackling these energy management challenges is essential to facilitate microgrid applications and seamlessly incorporate renewable energy resources. Artificial intelligence (AI) has recently demonstrated immense potential for optimizing energy management in microgrids, providing efficient and reliable solutions. This paper highlights the combined benefits of enabling AI-based methodologies in the energy management systems of microgrids by examining the applicability and efficiency of AI-based EMS in achieving specific technical and economic objectives. The paper also points out several future research directions that promise to spearhead AI-driven EMS, namely the development of self-healing microgrids, integration with blockchain technology, use of Internet of things (IoT), and addressing interpretability, data privacy, scalability, and the prospects to generative AI in the context of future AI-based EMS.

Meta federated learning (FL) is a personalized variant of FL, where multiple agents collaborate on training an initial shared model without exchanging raw dat a samples. The initial model should be trained in a way that current or new agents can easily adapt it to their local datasets after one or a few fine-tuning steps, thus improving the model personaliza tion. Conventional meta FL approaches minimize the average loss of agents on the local models obtai ned after one step of fine-tuning. In practice, agents may need to apply several fine-tuning steps to adapt the global model to their local data, especially under highly heterogeneous data dis tributions across agents. To this end, we present a generalized framework for the meta FL by minimizin g the average loss of agents on their local model after any arbitrary number ν of fine-tuning steps. For this generalized framework, we present a variant of the well-known federated averaging ( FedAvg) algorithm and conduct a comprehensive theoretical convergence analysis to charac terize the convergence speed as well as behavior of the meta loss functions in both the exact and appr oximated cases. Our experiments on real-world datasets demonstrate superior accuracy and fas ter convergence for the proposed scheme compared to conventional approaches.