This paper investigates the charging scheduling problem within an Electric Vehicle Charging Station (EVCS), aiming to coordinate multiple charging piles to optimize the charging process. The objectives are to increase the profit for the Charging Station Operator (CSO) and enhance user satisfaction while mitigating significant fluctuations in grid load. However, since these objectives originate from different stakeholders, they are inherently conflicting; a strategy trained using fixed objective weights cannot satisfy all of them simultaneously. Additionally, uncertainties such as vehicle arrival and departure times, power requirements, and grid tariffs complicate the modeling of this process. To address these challenges, this paper reformulates the scheduling process as a Markov Decision Process (MDP) and proposes a novel model-free reinforcement learning method. By integrating randomly varying objective preferences into the inputs of agents, a set of policies corresponding to different objective preferences can be learned through a single training session. As a result, agents can execute various policies to satisfy the stakeholders’ needs based on these preferences. Moreover, a sampling adjustment strategy is introduced to improve the uniformity of the final non-dominated solution set and enhance the ability of the agent to explore the objective space. Experiments show that the proposed method can effectively solve the multi-objective problem of cooperative charging within an EVCS. Compared to the baseline method, the proposed method can achieve superior results while simultaneously reducing the computational cost to a significant extent.

In recent years, Graph Neural Networks (GNNs) and Large Language Models (LLMs) have exhibited remarkable capability in addressing different graph learning and natural language tasks, respectively. Motivated by this, integrating LLMs with GNNs has been increasingly studied to acquire transferable knowledge across modalities, which leads to improved empirical performance in language and graph domains. However, existing studies mainly focused on a single-domain scenario by designing complicated integration techniques to manage multimodal data effectively. Therefore, a concise and generic learning framework for multi-domain tasks, i.e., graph and language domains, is highly desired yet remains under-exploited due to two major challenges. First, the language corpus of downstream tasks differs significantly from graph data, making it hard to bridge the knowledge gap between modalities. Second, not all knowledge demonstrates immediate benefits for downstream tasks, potentially introducing disruptive noise to context-sensitive models like LLMs. To tackle these challenges, we propose a novel plug-and-play framework for incorporating a lightweight cross-domain prompting method into both language and graph learning tasks. Specifically, we first convert the textual input into a domain-scalable prompt, which not only preserves the semantic and logical contents of the textual input, but also highlights related graph information as external knowledge for different domains. Then, we develop a reinforcement learning-based method to learn the optimal edge selection strategy for useful knowledge extraction, which profoundly sharpens the multi-domain model capabilities. In addition, we introduce a joint multi-view optimization module to regularize agent-level collaborative learning across two domains. Finally, extensive empirical justifications over 23 public and synthetic datasets demonstrate that our approach can be applied to diverse multi-domain tasks more accurately, robustly, and reasonably, and improve the performances of the state-of-the-art graph and language models in different learning paradigms.