Accurate motion prediction of surrounding agents is crucial for the safe planning of autonomous vehicles. Recent advancements have extended prediction techniques from individual agents to joint predictions of multiple interacting agents, with various strategies to address complex interactions within future motions of agents. However, these methods overlook the evolving nature of these interactions. To address this limitation, we propose a novel progressive multi-scale decoding strategy, termed ProgD, with the help of dynamic heterogeneous graph-based scenario modeling. In particular, to explicitly and comprehensively capture the evolving social interactions in future scenarios, given their inherent uncertainty, we design a progressive modeling of scenarios with dynamic heterogeneous graphs. With the unfolding of such dynamic heterogeneous graphs, a factorized architecture is designed to process the spatio-temporal dependencies within future scenarios and progressively eliminate uncertainty in future motions of multiple agents. Furthermore, a multi-scale decoding procedure is incorporated to improve on the future scenario modeling and consistent prediction of agents' future motion. Introduction Motion prediction is important for self-driving systems to ensure safe and efficient navigation. Of particular interest is joint multi-agent motion prediction, which involves concurrently forecasting the future trajectories of all agents within a scene. This task has gained increasing attention recently, due to its complexity compared to marginal motion prediction, as it requires maintaining consistency and coherence in future motions of interactive agents, reflecting the intricate dynamics of real-world traffic. Without such consistency, the prediction module could produce conflicting trajectories, such as collisions between the predicted motions of agents, which would undermine the reliability of the system and lead to unsafe or infeasible motion plans. The challenge is further compounded by several intrinsic factors: (1) Dynamic and complex future social interactions.

Constrained Reinforcement Learning (RL) aims to maximize the return while adhering to predefined constraint limits, which represent domain-specific safety requirements. In continuous control settings, where learning agents govern system actions, balancing the trade-off between reward maximization and constraint satisfaction remains a significant challenge. Policy optimization methods often exhibit instability near constraint boundaries, resulting in suboptimal training performance. To address this issue, we introduce a novel approach that integrates an adaptive incentive mechanism in addition to the reward structure to stay within the constraint bound before approaching the constraint boundary. Building on this insight, we propose Incrementally Penalized Proximal Policy Optimization (IP3O), a practical algorithm that enforces a progressively increasing penalty to stabilize training dynamics. Through empirical evaluation on benchmark environments, we demonstrate the efficacy of IP3O compared to the performance of state-of-the-art Safe RL algorithms. Furthermore, we provide theoretical guarantees by deriving a bound on the worst-case error of the optimality achieved by our algorithm.

Envy-freeness and the relaxation to Envy-freeness up to one item (EF-1) have been used as fairness concepts in the economics, game theory, and social choice literatures since the 1960s, and have recently gained popularity within the recommendation systems communities. In this short position paper we will give an overview of envy-freeness and its use in economics and recommendation systems; and illustrate why envy is not appropriate to measure fairness for use in settings where personalization plays a role.

The rapid advancement of large language models (LLMs) and domain-specific AI agents has greatly expanded the ecosystem of AI-powered services. User queries, however, are highly diverse and often span multiple domains and task types, resulting in a complex and heterogeneous landscape. This diversity presents a fundamental routing challenge: how to accurately direct each query to an appropriate execution unit while optimizing both performance and efficiency. To address this, we propose MoMA (Mixture of Models and Agents), a generalized routing framework that integrates both LLM and agent-based routing. Built upon a deep understanding of model and agent capabilities, MoMA effectively handles diverse queries through precise intent recognition and adaptive routing strategies, achieving an optimal balance between efficiency and cost. Specifically, we construct a detailed training dataset to profile the capabilities of various LLMs under different routing model structures, identifying the most suitable tasks for each LLM. During inference, queries are dynamically routed to the LLM with the best cost-performance efficiency. We also introduce an efficient agent selection strategy based on a context-aware state machine and dynamic masking. Experimental results demonstrate that the MoMA router offers superior cost-efficiency and scalability compared to existing approaches.

We present a preliminary experimental platform that explores how narrative elements might shape AI decision-making by combining reinforcement learning (RL) with language model reasoning. While AI systems can now both make decisions and engage in narrative reasoning, these capabilities have mostly been studied separately. Our platform attempts to bridge this gap using a dual-system architecture to examine how narrative frameworks could influence reward-based learning. The system comprises a reinforcement learning policy that suggests actions based on past experience, and a language model that processes these suggestions through different narrative frameworks to guide decisions. This setup enables initial experimentation with narrative elements while maintaining consistent environment and reward structures. We implement this architecture in a configurable gridworld environment, where agents receive both policy suggestions and information about their surroundings. The platform's modular design facilitates controlled testing of environmental complexity, narrative parameters, and the interaction between reinforcement learning and narrative-based decisions. Our logging system captures basic decision metrics, from RL policy values to language model reasoning to action selection patterns. While preliminary, this implementation provides a foundation for studying how different narrative frameworks might affect reward-based decisions and exploring potential interactions between optimization-based learning and symbolic reasoning in AI systems.

We study a problem of multi-agent exploration with behaviorally heterogeneous robots. Each robot maps its surroundings using SLAM and identifies a set of areas of interest (AoIs) or frontiers that are the most informative to explore next. The robots assess the utility of going to a frontier using Behavioral Entropy (BE) and then determine which frontier to go to via a distributed task assignment scheme. We convert the task assignment problem into a non-cooperative game and use a distributed algorithm (d-PBRAG) to converge to the Nash equilibrium (which we show is the optimal task allocation solution). For unknown utility cases, we provide robust bounds using approximate rewards. We test our algorithm (which has less communication cost and fast convergence) in simulation, where we explore the effect of sensing radii, sensing accuracy, and heterogeneity among robotic teams with respect to the time taken to complete exploration and path traveled. We observe that having a team of agents with heterogeneous behaviors is beneficial.

Multi-agent systems (MAS) are critical for automating complex tasks, yet their practical deployment is severely hampered by the challenge of failure attribution. Current diagnostic tools, which rely on statistical correlations, are fundamentally inadequate; on challenging benchmarks like Who\&When, state-of-the-art methods achieve less than 15\% accuracy in locating the root-cause step of a failure. To address this critical gap, we introduce the first failure attribution framework for MAS grounded in multi-granularity causal inference. Our approach makes two key technical contributions: (1) a performance causal inversion principle, which correctly models performance dependencies by reversing the data flow in execution logs, combined with Shapley values to accurately assign agent-level blame; (2) a novel causal discovery algorithm, CDC-MAS, that robustly identifies critical failure steps by tackling the non-stationary nature of MAS interaction data. The framework's attribution results directly fuel an automated optimization loop, generating targeted suggestions whose efficacy is validated via counterfactual simulations. Evaluations on the Who\&When and TRAIL benchmarks demonstrate a significant leap in performance. Our method achieves up to 36.2\% step-level accuracy. Crucially, the generated optimizations boost overall task success rates by an average of 22.4\%. This work provides a principled and effective solution for debugging complex agent interactions, paving the way for more reliable and interpretable multi-agent systems.

Post-training language models (LMs) with reinforcement learning (RL) can enhance their complex reasoning capabilities without supervised fine-tuning, as demonstrated by DeepSeek-R1-Zero. However, effectively utilizing RL for LMs requires significant parallelization to scale-up inference, which introduces non-trivial technical challenges (e.g. latency, memory, and reliability) alongside ever-growing financial costs. We present Swarm sAmpling Policy Optimization (SAPO), a fully decentralized and asynchronous RL post-training algorithm. SAPO is designed for decentralized networks of heterogenous compute nodes, where each node manages its own policy model(s) while "sharing" rollouts with others in the network; no explicit assumptions about latency, model homogeneity, or hardware are required and nodes can operate in silo if desired. As a result, the algorithm avoids common bottlenecks in scaling RL post-training while also allowing (and even encouraging) new possibilities. By sampling rollouts "shared" across the network, it enables "Aha moments" to propagate, thereby bootstrapping the learning process. In this paper we show SAPO achieved cumulative reward gains of up to 94% in controlled experiments. We also share insights from tests on a network with thousands of nodes contributed by Gensyn community members running the algorithm on diverse hardware and models during an open-source demo.

Autonomous-driving research has recently embraced deep Reinforcement Learning (RL) as a promising framework for data-driven decision making, yet a clear picture of how these algorithms are currently employed, benchmarked and evaluated is still missing. This survey fills that gap by systematically analysing around 100 peer-reviewed papers that train, test or validate RL policies inside the open-source CARLA simulator. We first categorize the literature by algorithmic family model-free, model-based, hierarchical, and hybrid and quantify their prevalence, highlighting that more than 80% of existing studies still rely on model-free methods such as DQN, PPO and SAC. Next, we explain the diverse state, action and reward formulations adopted across works, illustrating how choices of sensor modality (RGB, LiDAR, BEV, semantic maps, and carla kinematics states), control abstraction (discrete vs. continuous) and reward shaping are used across various literature. We also consolidate the evaluation landscape by listing the most common metrics (success rate, collision rate, lane deviation, driving score) and the towns, scenarios and traffic configurations used in CARLA benchmarks. Persistent challenges including sparse rewards, sim-to-real transfer, safety guarantees and limited behaviour diversity are distilled into a set of open research questions, and promising directions such as model-based RL, meta-learning and richer multi-agent simulations are outlined. By providing a unified taxonomy, quantitative statistics and a critical discussion of limitations, this review aims to serve both as a reference for newcomers and as a roadmap for advancing RL-based autonomous driving toward real-world deployment.

In a robotic swarm, parameters such as position and proximity to the target can be described in terms of probability amplitudes. This idea led to recent studies on a quantum approach to the definition of the swarm, including a block-matrix representation. Here, we propose an advancement of the idea, defining a swarm as a mixed quantum state, to be described with a density matrix, whose size does not change with the number of robots. We end the article with some directions for future research.