Signal control in urban corridors faces the dual challenge of maintaining arterial traffic progression while adapting to demand variations at local intersections. We propose a hierarchical traffic signal coordination and control scheme that integrates model-based optimization with reinforcement learning. The system consists of: (i) a High-Level Coordinator (HLC) that selects coordination strategies based on observed and predicted demand; (ii) a Corridor Coordinator that derives phase constraints from the selected strategy-either Max-Flow Coordination (MFC) or Green-Wave Coordination (GWC); and (iii) Hybrid Signal Agents (HSAs) that determine signal phases via reinforcement learning with action masking to enforce feasibility. Hierarchical reinforcement learning with Proximal Policy Optimization (PPO) is used to train HSA and HLC policies. At the lower level, three HSA policies-MFC-aware, GWC-aware, and pure agent control (PAC) are trained in conjunction with their respective coordination strategies. At the higher level, the HLC is trained to dynamically switch strategies using a multi-objective reward balancing corridor-level and network-wide performance. The proposed scheme was developed and evaluated on a SUMO-RLlib platform. Case results show that hybrid MFC maximizes throughput under heavy demand; hybrid GWC consistently minimizes arterial stops and maintains progression across diverse traffic conditions but can reduce network-wide efficiency; and PAC improves network-wide travel time in moderate demand but is less effective under heavy demand. The hierarchical design enables adaptive strategy selection, achieving robust performance across all demand levels.

Large language models are extensively utilized in creative writing applications. Creative writing requires a balance between subjective writing quality (e.g., literariness and emotional expression) and objective constraint following (e.g., format requirements and word limits). Existing methods find it difficult to balance these two aspects: single reward strategies fail to improve both abilities simultaneously, while fixed-weight mixed-reward methods lack the ability to adapt to different writing scenarios. To address this problem, we propose Reinforcement Learning with Mixed Rewards (RLMR), utilizing a dynamically mixed reward system from a writing reward model evaluating subjective writing quality and a constraint verification model assessing objective constraint following. The constraint following reward weight is adjusted dynamically according to the writing quality within sampled groups, ensuring that samples violating constraints get negative advantage in GRPO and thus penalized during training, which is the key innovation of this proposed method. We conduct automated and manual evaluations across diverse model families from 8B to 72B parameters. Additionally, we construct a real-world writing benchmark named WriteEval for comprehensive evaluation. Results illustrate that our method achieves consistent improvements in both instruction following (IFEval from 83.36% to 86.65%) and writing quality (72.75% win rate in manual expert pairwise evaluations on WriteEval). To the best of our knowledge, RLMR is the first work to combine subjective preferences with objective verification in online RL training, providing an effective solution for multi-dimensional creative writing optimization.

Risk-averse Constrained Reinforcement Learning (RaCRL) aims to learn policies that minimise the likelihood of rare and catastrophic constraint violations caused by an environment's inherent randomness. In general, risk-aversion leads to conservative exploration of the environment which typically results in converging to sub-optimal policies that fail to adequately maximise reward or, in some cases, fail to achieve the goal. In this paper, we propose an exploration-based approach for RaCRL called Optimistic Risk-averse Actor Critic (ORAC), which constructs an exploratory policy by maximising a local upper confidence bound of the state-action reward value function whilst minimising a local lower confidence bound of the risk-averse state-action cost value function. Specifically, at each step, the weighting assigned to the cost value is increased or decreased if it exceeds or falls below the safety constraint value. This way the policy is encouraged to explore uncertain regions of the environment to discover high reward states whilst still satisfying the safety constraints. Our experimental results demonstrate that the ORAC approach prevents convergence to sub-optimal policies and improves significantly the reward-cost trade-off in various continuous control tasks such as Safety-Gymnasium and a complex building energy management environment CityLearn.

-- In swarm robotics, confrontation scenarios, including strategic confrontations, require efficient decision-making that integrates discrete commands and continuous actions. Traditional task and motion planning methods separate decision-making into two layers, but their unidirectional structure fails to capture the interdependence between these layers, limiting adaptability in dynamic environments. Here, we propose a novel bidirectional approach based on hierarchical reinforcement learning, enabling dynamic interaction between the layers. This method effectively maps commands to task allocation and actions to path planning, while leveraging cross-training techniques to enhance learning across the hierarchical framework. Furthermore, we introduce a trajectory prediction model that bridges abstract task representations with actionable planning goals. In our experiments, it achieves over 80% in confrontation win rate and under 0.01 seconds in decision time, outperforming existing approaches. Demonstrations through large-scale tests and real-world robot experiments further emphasize the generalization capabilities and practical applicability of our method. I. INTRODUCTION Recent advances in artificial intelligence lead to significant progress in robotics [1], [2], with particular attention given to robotic swarm confrontations [3], [4].

Autonomous agents for Graphical User Interfaces (GUIs) face significant challenges in specialized domains such as scientific computing, where both long-horizon planning and precise execution are required. Existing approaches suffer from a trade-off: generalist agents excel at planning but perform poorly in execution, while specialized agents demonstrate the opposite weakness. Recent compositional frameworks attempt to bridge this gap by combining a planner and an actor, but they are typically static and non-trainable, which prevents adaptation from experience. This is a critical limitation given the scarcity of high-quality data in scientific domains. To address these limitations, we introduce CODA, a novel and trainable compositional framework that integrates a generalist planner (Cerebrum) with a specialist executor (Cerebellum), trained via a dedicated two-stage pipeline. In the first stage, Specialization, we apply a decoupled GRPO approach to train an expert planner for each scientific application individually, bootstrapping from a small set of task trajectories. In the second stage, Generalization, we aggregate all successful trajectories from the specialized experts to build a consolidated dataset, which is then used for supervised fine-tuning of the final planner. This equips CODA with both robust execution and cross-domain generalization. Evaluated on four challenging applications from the ScienceBoard benchmark, CODA significantly outperforms baselines and establishes a new state of the art among open-source models.

Failure-Directed Search (FDS) is a significant complete generic search algorithm used in Constraint Programming (CP) to efficiently explore the search space, proven particularly effective on scheduling problems. This paper analyzes FDS's properties, showing that minimizing the size of its search tree guided by ranked branching decisions is closely related to the Multi-armed bandit (MAB) problem. Building on this insight, MAB reinforcement learning algorithms are applied to FDS, extended with problem-specific refinements and parameter tuning, and evaluated on the two most fundamental scheduling problems, the Job Shop Scheduling Problem (JSSP) and Resource-Constrained Project Scheduling Problem (RCPSP). The resulting enhanced FDS, using the best extended MAB algorithm and configuration, performs 1.7 times faster on the JSSP and 2.1 times faster on the RCPSP benchmarks compared to the original implementation in a new solver called OptalCP, while also being 3.5 times faster on the JSSP and 2.1 times faster on the RCPSP benchmarks than the current state-of-the-art FDS algorithm in IBM CP Optimizer 22.1. Furthermore, using only a 900-second time limit per instance, the enhanced FDS improved the existing state-of-the-art lower bounds of 78 of 84 JSSP and 226 of 393 RCPSP standard open benchmark instances while also completely closing a few of them.

Offline reinforcement learning (RL) enables learning effective policies from fixed datasets without any environment interaction. Existing methods typically employ policy constraints to mitigate the distribution shift encountered during offline RL training. However, because the scale of the constraints varies across tasks and datasets of differing quality, existing methods must meticulously tune hyperparameters to match each dataset, which is time-consuming and often impractical. We propose Adaptive Scaling of Policy Constraints (ASPC), a second-order differentiable framework that dynamically balances RL and behavior cloning (BC) during training. We theoretically analyze its performance improvement guarantee. In experiments on 39 datasets across four D4RL domains, ASPC using a single hyperparameter configuration outperforms other adaptive constraint methods and state-of-the-art offline RL algorithms that require per-dataset tuning while incurring only minimal computational overhead. The code will be released at https://github.com/Colin-Jing/ASPC.

The suggested procedure involves integrating the proximity function null P as a mechanism to guide exploration on the space of walks. While the use of null P that we have explained has focused on classification and metric analysis, its geometric interpretation and ability to quantify similarity between walks suggest a broader applicability, particularly in settings where the reward landscape is sparse or the graph structure is too large for exhaustive exploration. We propose an improvement to the exploration strategy used in reinforcement learning algorithms that incrementally construct walks within graph-based environments. Traditionally, these algorithms alternate between exploitation (choosing the next node to maximize an estimated reward) and exploration (randomly selecting a new node). The novelty lies in replacing random exploration with a proximity-guided strategy using a function null P . Instead of sampling uniformly, the agent compares potential path extensions to a reference set of high-reward walks, prioritizing those that are most similar in structure. This approach introduces a more informed, data-driven method for exploration, focusing on areas of the graph that resemble previously successful trajectories.

This thesis examines the obstacles and potential solutions for creating such bots, focusing on innovative techniques that enable bots to learn and adapt autonomously in constantly changing environments. End-to-end task bots, typically built using a static and limited corpus, face difficulties when deployed online due to three primary factors tied to this limitation. First, they might confront queries featuring unexpected linguistic patterns or slot values (i.e., unseen user behaviors). Second, they could potentially face requirements for new functions or tasks (i.e., task definition extensions). Third, even when equipped with relevant knowledge, these bots may produce responses that appear plausible but are actually incorrect (i.e., "hallucinations"). Addressing these challenges is vital for enhancing task bots' performance and reliability in real-world settings.

This paper presents an Impedance Primitive-augmented hierarchical reinforcement learning framework for efficient robotic manipulation in sequential contact tasks. We leverage this hierarchical structure to sequentially execute behavior primitives with variable stiffness control capabilities for contact tasks. Our proposed approach relies on three key components: an action space enabling variable stiffness control, an adaptive stiffness controller for dynamic stiffness adjustments during primitive execution, and affordance coupling for efficient exploration while encouraging compliance. Through comprehensive training and evaluation, our framework learns efficient stiffness control capabilities and demonstrates improvements in learning efficiency, compositionality in primitive selection, and success rates compared to the state-of-the-art. The training environments include block lifting, door opening, object pushing, and surface cleaning. Real world evaluations further confirm the framework's sim2real capability. This work lays the foundation for more adaptive and versatile robotic manipulation systems, with potential applications in more complex contact-based tasks.