The process of creating training data to teach models is currently driven by humans, who manually analyze model weaknesses and plan how to create data that improves a student model. Approaches using LLMs as annotators reduce human effort, but still require humans to interpret feedback from evaluations and control the LLM to produce data the student needs. Automating this labor-intensive process by creating autonomous data generation agents - or teachers - is desirable, but requires environments that can simulate the feedback-driven, iterative, closed loop of data creation. To enable rapid, scalable testing for such agents and their modules, we introduce DataEnvGym, a testbed of teacher environments for data generation agents. DataEnvGym frames data generation as a sequential decision-making task, involving an agent consisting of a data generation policy (which generates a plan for creating training data) and a data generation engine (which transforms the plan into data), inside an environment that provides student feedback. The agent's goal is to improve student performance. Students are iteratively trained and evaluated on generated data, and their feedback (in the form of errors or weak skills) is reported to the agent after each iteration. DataEnvGym includes multiple teacher environment instantiations across 3 levels of structure in the state representation and action space. More structured environments are based on inferred skills and offer more interpretability and curriculum control. We support 4 domains (math, code, VQA, and tool-use) and test multiple students and teachers. Example agents in our teaching environments can iteratively improve students across tasks and settings. Moreover, we show that environments teach different skill levels and test variants of key modules, pointing to future work in improving data generation agents, engines, and feedback mechanisms.

Multi-Robot Coverage problems have been extensively studied in robotics, planning and multi-agent systems. In this work, we consider the coverage problem when there are constraints on the proximity (e.g., maximum distance between the agents, or a blue agent must be adjacent to a red agent) and the movement (e.g., terrain traversability and material load capacity) of the robots. Such constraints naturally arise in many real-world applications, e.g. in search-and-rescue and maintenance operations. Given such a setting, the goal is to compute a covering tour of the graph with a minimum number of steps, and that adheres to the proximity and movement constraints. For this problem, our contributions are four: (i) a formal formulation of the problem, (ii) an exact algorithm that is FPT in F, d and tw, the set of robot formations that encode the proximity constraints, the maximum nodes degree, and the tree-width of the graph, respectively, (iii) for the case that the graph is a tree: a PTAS approximation scheme, that given an approximation parameter epsilon, produces a tour that is within a epsilon times error(||F||, d) of the optimal one, and the computation runs in time poly(n) times h(1/epsilon,||F||). (iv) for the case that the graph is a tree, with $k=3$ robots, and the constraint is that all agents are connected: a PTAS scheme with multiplicative approximation error of 1+O(epsilon), independent of the maximal degree d.

The problem of decentralized multi-robot patrol has previously been approached primarily with hand-designed strategies for minimization of 'idlenes' over the vertices of a graph-structured environment. Here we present two lightweight neural network-based strategies to tackle this problem, and show that they significantly outperform existing strategies in both idleness minimization and against an intelligent intruder model, as well as presenting an examination of robustness to communication failure. Our results also indicate important considerations for future strategy design.

For embodied reinforcement learning (RL) agents interacting with the environment, it is desirable to have rapid policy adaptation to unseen visual observations, but achieving zero-shot adaptation capability is considered as a challenging problem in the RL context. To address the problem, we present a novel contrastive prompt ensemble (ConPE) framework which utilizes a pretrained vision-language model and a set of visual prompts, thus enabling efficient policy learning and adaptation upon a wide range of environmental and physical changes encountered by embodied agents. Specifically, we devise a guided-attention-based ensemble approach with multiple visual prompts on the vision-language model to construct robust state representations. Each prompt is contrastively learned in terms of an individual domain factor that significantly affects the agent's egocentric perception and observation. For a given task, the attention-based ensemble and policy are jointly learned so that the resulting state representations not only generalize to various domains but are also optimized for learning the task. Through experiments, we show that ConPE outperforms other state-of-the-art algorithms for several embodied agent tasks including navigation in AI2THOR, manipulation in egocentric-Metaworld, and autonomous driving in CARLA, while also improving the sample efficiency of policy learning and adaptation.

In human society, the conflict between self-interest and collective well-being often obstructs efforts to achieve shared welfare. Related concepts like the Tragedy of the Commons and Social Dilemmas frequently manifest in our daily lives. As artificial agents increasingly serve as autonomous proxies for humans, we propose using multi-agent reinforcement learning (MARL) to address this issue - learning policies to maximise collective returns even when individual agents' interests conflict with the collective one. Traditional MARL solutions involve sharing rewards, values, and policies or designing intrinsic rewards to encourage agents to learn collectively optimal policies. We introduce a novel MARL approach based on Suggestion Sharing (SS), where agents exchange only action suggestions. This method enables effective cooperation without the need to design intrinsic rewards, achieving strong performance while revealing less private information compared to sharing rewards, values, or policies. Our theoretical analysis establishes a bound on the discrepancy between collective and individual objectives, demonstrating how sharing suggestions can align agents' behaviours with the collective objective. Experimental results demonstrate that SS performs competitively with baselines that rely on value or policy sharing or intrinsic rewards.

Large Language Models (LLMs) have demonstrated remarkable performance across various tasks. A promising but largely under-explored area is their potential to facilitate human coordination with many agents. Such capabilities would be useful in domains including disaster response, urban planning, and real-time strategy scenarios. In this work, we introduce (1) a real-time strategy game benchmark designed to evaluate these abilities and (2) a novel framework we term HIVE. HIVE empowers a single human to coordinate swarms of up to 2,000 agents using natural language dialog with an LLM. We present promising results on this multi-agent benchmark, with our hybrid approach solving tasks such as coordinating agent movements, exploiting unit weaknesses, leveraging human annotations, and understanding terrain and strategic points. However, our findings also highlight critical limitations of current models, including difficulties in processing spatial visual information and challenges in formulating long-term strategic plans. This work sheds light on the potential and limitations of LLMs in human-swarm coordination, paving the way for future research in this area. The HIVE project page, which includes videos of the system in action, can be found here: hive.syrkis.com.

Additionally, the investigation into the nested pointwise aggregation of In the realm of real-time online Gaussian Process (GP) experts has been undertaken [20], [21]. Nevertheless, the regression, continuously collecting the training data becomes application of pointwise aggregation across the entirety of impractical for dynamic systems due to the constraints in the training dataset proves unattainable within distributed physical storage space and the escalating computational burden systems. Instead of employing the entire dataset for prediction, poses substantial practical challenges, particularly in real-time several approximation techniques prove instrumental. Moreover, local approximation B. Agent-based Gaussian Process methods, such as the naive local experts, the mixture of Distributed learning finds prominent application in multiagent experts, and the product of experts, present viable alternatives. Consequently, joint predictions are aggregated [8]. Several efforts have been dedicated to implementing distributed Prominently, cooperative learning within distributed systems Gaussian Process (DGP) methodologies within MASs.

In real world software development, improper or missing exception handling can severely impact the robustness and reliability of code. Exception handling mechanisms require developers to detect, capture, and manage exceptions according to high standards, but many developers struggle with these tasks, leading to fragile code. This problem is particularly evident in open-source projects and impacts the overall quality of the software ecosystem. To address this challenge, we explore the use of large language models (LLMs) to improve exception handling in code. Through extensive analysis, we identify three key issues: Insensitive Detection of Fragile Code, Inaccurate Capture of Exception Block, and Distorted Handling Solution. These problems are widespread across real world repositories, suggesting that robust exception handling practices are often overlooked or mishandled. In response, we propose Seeker, a multi-agent framework inspired by expert developer strategies for exception handling. Seeker uses agents: Scanner, Detector, Predator, Ranker, and Handler to assist LLMs in detecting, capturing, and resolving exceptions more effectively. Our work is the first systematic study on leveraging LLMs to enhance exception handling practices in real development scenarios, providing valuable insights for future improvements in code reliability.

Recent advances in multi-agent systems manipulation have demonstrated a rising demand for the implementation of multi-UAV systems in urban areas which are always subjected to the presence of static and dynamic obstacles. The focus of the presented research is on the introduction of a nature-inspired collision-free control for a multi-UAV system considering obstacle avoidance maneuvers. Inspired by the collective behavior of tilapia fish and pigeon, the presented framework in this study uses a centralized controller for the optimal formation control/recovery, which is defined by probabilistic Lloyd's algorithm, while it uses a distributed controller for the intervehicle collision and obstacle avoidance. Further, the presented framework has been extended to the 3D space with 3D maneuvers. Finally, the presented framework has been applied to a multi-UAV system in 2D and 3D scenarios, and obtained results demonstrated the validity of the presented method in the presence of buildings and different types of obstacles. Keywords: Multi-Agent System, Obstacle Avoidance, Collision Avoidance, Formation Control, Centroidal Voronoi Tessellation, Distributed Control.

Collaborative Perception (CP) has shown a promising technique for autonomous driving, where multiple connected and autonomous vehicles (CAVs) share their perception information to enhance the overall perception performance and expand the perception range. However, in CP, ego CAV needs to receive messages from its collaborators, which makes it easy to be attacked by malicious agents. For example, a malicious agent can send harmful information to the ego CAV to mislead it. To address this critical issue, we propose a novel method, \textbf{CP-Guard}, a tailored defense mechanism for CP that can be deployed by each agent to accurately detect and eliminate malicious agents in its collaboration network. Our key idea is to enable CP to reach a consensus rather than a conflict against the ego CAV's perception results. Based on this idea, we first develop a probability-agnostic sample consensus (PASAC) method to effectively sample a subset of the collaborators and verify the consensus without prior probabilities of malicious agents. Furthermore, we define a collaborative consistency loss (CCLoss) to capture the discrepancy between the ego CAV and its collaborators, which is used as a verification criterion for consensus. Finally, we conduct extensive experiments in collaborative bird's eye view (BEV) tasks and our results demonstrate the effectiveness of our CP-Guard.