Minrui Xu

Minrui Xu, Dusit Niyato, Junlong Chen et al.

In the vehicular mixed reality (MR) Metaverse, the distance between physical and virtual entities can be overcome by fusing the physical and virtual environments with multi-dimensional communications in autonomous driving systems. Assisted by digital twin (DT) technologies, connected autonomous vehicles (AVs), roadside units (RSU), and virtual simulators can maintain the vehicular MR Metaverse via digital simulations for sharing data and making driving decisions collaboratively. However, large-scale traffic and driving simulation via realistic data collection and fusion from the physical world for online prediction and offline training in autonomous driving systems are difficult and costly. In this paper, we propose an autonomous driving architecture, where generative AI is leveraged to synthesize unlimited conditioned traffic and driving data in simulations for improving driving safety and traffic efficiency. First, we propose a multi-task DT offloading model for the reliable execution of heterogeneous DT tasks with different requirements at RSUs. Then, based on the preferences of AV's DTs and collected realistic data, virtual simulators can synthesize unlimited conditioned driving and traffic datasets to further improve robustness. Finally, we propose a multi-task enhanced auction-based mechanism to provide fine-grained incentives for RSUs in providing resources for autonomous driving. The property analysis and experimental results demonstrate that the proposed mechanism and architecture are strategy-proof and effective, respectively.

Minrui Xu, Xiaoxu Ren, Dusit Niyato et al.

With the drive to create a decentralized digital economy, Web 3.0 has become a cornerstone of digital transformation, developed on the basis of computing-force networking, distributed data storage, and blockchain. With the rapid realization of quantum devices, Web 3.0 is being developed in parallel with the deployment of quantum cloud computing and quantum Internet. In this regard, quantum computing first disrupts the original cryptographic systems that protect data security while reshaping modern cryptography with the advantages of quantum computing and communication. Therefore, in this paper, we introduce a quantum blockchain-driven Web 3.0 framework that provides information-theoretic security for decentralized data transferring and payment transactions. First, we present the framework of quantum blockchain-driven Web 3.0 with future-proof security during the transmission of data and transaction information. Next, we discuss the potential applications and challenges of implementing quantum blockchain in Web 3.0. Finally, we describe a use case for quantum non-fungible tokens (NFTs) and propose a quantum deep learning-based optimal auction for NFT trading to maximize the achievable revenue for sufficient liquidity in Web 3.0. In this way, the proposed framework can achieve proven security and sustainability for the next-generation decentralized digital society.

Chao Ren, Rudai Yan, Huihui Zhu et al.

Quantum Federated Learning (QFL) is an emerging interdisciplinary field that merges the principles of Quantum Computing (QC) and Federated Learning (FL), with the goal of leveraging quantum technologies to enhance privacy, security, and efficiency in the learning process. Currently, there is no comprehensive survey for this interdisciplinary field. This review offers a thorough, holistic examination of QFL. We aim to provide a comprehensive understanding of the principles, techniques, and emerging applications of QFL. We discuss the current state of research in this rapidly evolving field, identify challenges and opportunities associated with integrating these technologies, and outline future directions and open research questions. We propose a unique taxonomy of QFL techniques, categorized according to their characteristics and the quantum techniques employed. As the field of QFL continues to progress, we can anticipate further breakthroughs and applications across various industries, driving innovation and addressing challenges related to data privacy, security, and resource optimization. This review serves as a first-of-its-kind comprehensive guide for researchers and practitioners interested in understanding and advancing the field of QFL.

Minrui Xu, Dusit Niyato, Hongliang Zhang et al.

Metaverse seamlessly blends the physical world and virtual space via ubiquitous communication and computing infrastructure. In transportation systems, the vehicular Metaverse can provide a fully-immersive and hyperreal traveling experience (e.g., via augmented reality head-up displays, AR-HUDs) to drivers and users in autonomous vehicles (AVs) via roadside units (RSUs). However, provisioning real-time and immersive services necessitates effective physical-virtual synchronization between physical and virtual entities, i.e., AVs and Metaverse AR recommenders (MARs). In this paper, we propose a generative AI-empowered physical-virtual synchronization framework for the vehicular Metaverse. In physical-to-virtual synchronization, digital twin (DT) tasks generated by AVs are offloaded for execution in RSU with future route generation. In virtual-to-physical synchronization, MARs customize diverse and personal AR recommendations via generative AI models based on user preferences. Furthermore, we propose a multi-task enhanced auction-based mechanism to match and price AVs and MARs for RSUs to provision real-time and effective services. Finally, property analysis and experimental results demonstrate that the proposed mechanism is strategy-proof and adverse-selection free while increasing social surplus by 50%.

Junlong Chen, Jiawen Kang, Minrui Xu et al.

Avatars, as promising digital assistants in Vehicular Metaverses, can enable drivers and passengers to immerse in 3D virtual spaces, serving as a practical emerging example of Artificial Intelligence of Things (AIoT) in intelligent vehicular environments. The immersive experience is achieved through seamless human-avatar interaction, e.g., augmented reality navigation, which requires intensive resources that are inefficient and impractical to process on intelligent vehicles locally. Fortunately, offloading avatar tasks to RoadSide Units (RSUs) or cloud servers for remote execution can effectively reduce resource consumption. However, the high mobility of vehicles, the dynamic workload of RSUs, and the heterogeneity of RSUs pose novel challenges to making avatar migration decisions. To address these challenges, in this paper, we propose a dynamic migration framework for avatar tasks based on real-time trajectory prediction and Multi-Agent Deep Reinforcement Learning (MADRL). Specifically, we propose a model to predict the future trajectories of intelligent vehicles based on their historical data, indicating the future workloads of RSUs.Based on the expected workloads of RSUs, we formulate the avatar task migration problem as a long-term mixed integer programming problem. To tackle this problem efficiently, the problem is transformed into a Partially Observable Markov Decision Process (POMDP) and solved by multiple DRL agents with hybrid continuous and discrete actions in decentralized. Numerical results demonstrate that our proposed algorithm can effectively reduce the latency of executing avatar tasks by around 25% without prediction and 30% with prediction and enhance user immersive experiences in the AIoT-enabled Vehicular Metaverse (AeVeM).

Siyue Zhang, Minrui Xu, Wei Yang Bryan Lim et al.

Recent breakthroughs in generative artificial intelligence have triggered a surge in demand for machine learning training, which poses significant cost burdens and environmental challenges due to its substantial energy consumption. Scheduling training jobs among geographically distributed cloud data centers unveils the opportunity to optimize the usage of computing capacity powered by inexpensive and low-carbon energy and address the issue of workload imbalance. To tackle the challenge of multi-objective scheduling, i.e., maximizing GPU utilization while reducing operational costs, we propose an algorithm based on multi-agent reinforcement learning and actor-critic methods to learn the optimal collaborative scheduling strategy through interacting with a cloud system built with real-life workload patterns, energy prices, and carbon intensities. Compared with other algorithms, our proposed method improves the system utility by up to 28.6% attributable to higher GPU utilization, lower energy cost, and less carbon emission.

Yahao Ding, Wen Shang, Minrui Xu et al.

The Metaverse, a burgeoning collective virtual space merging augmented reality and persistent virtual worlds, necessitates advanced artificial intelligence (AI) and communication technologies to support immersive and interactive experiences. Federated learning (FL) has emerged as a promising technique for collaboratively training AI models while preserving data privacy. However, FL faces challenges such as high communication overhead and substantial computational demands, particularly for neural network (NN) models. To address these issues, we propose an integrated federated split learning and hyperdimensional computing (FSL-HDC) framework for emerging foundation models. This novel approach reduces communication costs, computation load, and privacy risks, making it particularly suitable for resource-constrained edge devices in the Metaverse, ensuring real-time responsive interactions. Additionally, we introduce an optimization algorithm that concurrently optimizes transmission power and bandwidth to minimize the maximum transmission time among all users to the server. The simulation results based on the MNIST dataset indicate that FSL-HDC achieves an accuracy rate of approximately 87.5%, which is slightly lower than that of FL-HDC. However, FSL-HDC exhibits a significantly faster convergence speed, approximately 3.733x that of FSL-NN, and demonstrates robustness to non-IID data distributions. Moreover, our proposed optimization algorithm can reduce the maximum transmission time by up to 64% compared with the baseline.

Qianzhou Chen, Siqi Sun, Minrui Xu et al.

The rapid advancements in foundation models and sixth-generation (6G) wireless communication systems necessitate the development of efficient, scalable, and privacy-preserving machine learning approaches. For foundation models in 6G, split learning (SL) and aggregation learning (AL) have emerged as promising paradigms that address key challenges in distributed artificial intelligence (AI), such as communication efficiency, resource allocation, and data privacy. SL enables multiple entities to collaboratively train deep learning models by partitioning neural networks, while AL focuses on aggregating intermediate results or model updates from multiple participants, improving robustness, optimizing resource utilization, and mitigating data leakage risks. Specifically, SL is ideal for scenarios requiring strict data isolation (e.g., vertical collaborations), whereas AL suits homogeneous horizontal data settings; they can be combined to balance privacy and communication efficiency. This survey provides a comprehensive analysis of SL and AL in 6G communication systems, exploring their architectures, technical methodologies, and integration with AI-native 6G communication technologies. We examine different SL configurations, aggregation techniques, and their roles in optimizing distributed foundation models. Furthermore, we discuss their applications in emerging wireless networks, including semantic communication, reconfigurable intelligent surfaces (RIS), space-air-ground integrated networks (SAGINs), and quantum communication. By analyzing the impact of SL and AL, this survey provides insights into their role in shaping distributed AI-driven communication systems in the 6G era, focusing on efficiency, privacy preservation, and scalability.

KaShun Shum, Minrui Xu, Jianshu Zhang et al.

Large language models (LLMs) have become increasingly prevalent in our daily lives, leading to an expectation for LLMs to be trustworthy -- - both accurate and well-calibrated (the prediction confidence should align with its ground truth correctness likelihood). Nowadays, fine-tuning has become the most popular method for adapting a model to practical usage by significantly increasing accuracy on downstream tasks. Despite the great accuracy it achieves, we found fine-tuning is still far away from satisfactory trustworthiness due to "tuning-induced mis-calibration". In this paper, we delve deeply into why and how mis-calibration exists in fine-tuned models, and how distillation can alleviate the issue. Then we further propose a brand new method named Efficient Trustworthy Distillation (FIRST), which utilizes a small portion of teacher's knowledge to obtain a reliable language model in a cost-efficient way. Specifically, we identify the "concentrated knowledge" phenomenon during distillation, which can significantly reduce the computational burden. Then we apply a "trustworthy maximization" process to optimize the utilization of this small portion of concentrated knowledge before transferring it to the student. Experimental results demonstrate the effectiveness of our method, where better accuracy (+2.3%) and less mis-calibration (-10%) are achieved on average across both in-domain and out-of-domain scenarios, indicating better trustworthiness.

Chengbo Zheng, Yuanhao Zhang, Zeyu Huang et al.

Interdisciplinary studies often require researchers to explore literature in diverse branches of knowledge. Yet, navigating through the highly scattered knowledge from unfamiliar disciplines poses a significant challenge. In this paper, we introduce DiscipLink, a novel interactive system that facilitates collaboration between researchers and large language models (LLMs) in interdisciplinary information seeking (IIS). Based on users' topics of interest, DiscipLink initiates exploratory questions from the perspectives of possible relevant fields of study, and users can further tailor these questions. DiscipLink then supports users in searching and screening papers under selected questions by automatically expanding queries with disciplinary-specific terminologies, extracting themes from retrieved papers, and highlighting the connections between papers and questions. Our evaluation, comprising a within-subject comparative experiment and an open-ended exploratory study, reveals that DiscipLink can effectively support researchers in breaking down disciplinary boundaries and integrating scattered knowledge in diverse fields. The findings underscore the potential of LLM-powered tools in fostering information-seeking practices and bolstering interdisciplinary research.

Minrui Xu, Zilin Wang, Mengyi DENG et al.

Equipping LLMs with tool-use capabilities via Agentic Reinforcement Learning (Agentic RL) is bottlenecked by two challenges: the lack of scalable, robust execution environments and the scarcity of realistic training data that captures implicit human reasoning. Existing approaches depend on costly real-world APIs, hallucination-prone LLM simulators, or synthetic environments that are often single-turn or depend on pre-collected documents. Moreover, synthetic trajectories are frequently over-specified, resembling instruction sequences rather than natural human intents, reducing their effectiveness for RL training. We introduce EnvFactory, a fully automated framework that addresses both challenges. EnvFactory autonomously explores and verifies stateful, executable tool environments from authentic resources, and synthesizes natural multi-turn trajectories through topology-aware sampling and calibrated refinement, producing grounded queries with implicit intents. Using only 85 verified environments across 7 domains, EnvFactory generates 2,575 SFT and RL trajectories. Despite using significantly fewer environments than prior work, which are often 5 times more, EnvFactory achieves superior training efficiency and downstream performance, improving Qwen3-series models by up to +15% on BFCLv3, +8.6% on MCP-Atlas, and +6% on conversational benchmarks including $τ^2$-Bench and VitaBench. By fully automating both environment construction and trajectory synthesis, EnvFactory provides a scalable, extensible, and robust foundation for Agentic RL.

Xu Yang, Xiao Yang, Shikai Fang et al.

Recent advances in AI and ML have transformed data science, yet increasing complexity and expertise requirements continue to hinder progress. Although crowd-sourcing platforms alleviate some challenges, high-level machine learning engineering (MLE) tasks remain labor-intensive and iterative. We introduce R&D-Agent, a comprehensive, decoupled, and extensible framework that formalizes the MLE process. R&D-Agent defines the MLE workflow into two phases and six components, turning agent design for MLE from ad-hoc craftsmanship into a principled, testable process. Although several existing agents report promising gains on their chosen components, they can mostly be summarized as a partial optimization from our framework's simple baseline. Inspired by human experts, we designed efficient and effective agents within this framework that achieve state-of-the-art performance. Evaluated on MLE-Bench, the agent built on R&D-Agent ranks as the top-performing machine learning engineering agent, achieving 35.1% any medal rate, demonstrating the ability of the framework to speed up innovation and improve accuracy across a wide range of data science applications. We have open-sourced R&D-Agent on GitHub: https://github.com/microsoft/RD-Agent.

Yuante Li, Xu Yang, Xiao Yang et al.

Financial markets pose fundamental challenges for asset return prediction due to their high dimensionality, non-stationarity, and persistent volatility. Despite advances in large language models and multi-agent systems, current quantitative research pipelines suffer from limited automation, weak interpretability, and fragmented coordination across key components such as factor mining and model innovation. In this paper, we propose R&D-Agent for Quantitative Finance, in short RD-Agent(Q), the first data-centric multi-agent framework designed to automate the full-stack research and development of quantitative strategies via coordinated factor-model co-optimization. RD-Agent(Q) decomposes the quant process into two iterative stages: a Research stage that dynamically sets goal-aligned prompts, formulates hypotheses based on domain priors, and maps them to concrete tasks, and a Development stage that employs a code-generation agent, Co-STEER, to implement task-specific code, which is then executed in real-market backtests. The two stages are connected through a feedback stage that thoroughly evaluates experimental outcomes and informs subsequent iterations, with a multi-armed bandit scheduler for adaptive direction selection. Empirically, RD-Agent(Q) achieves up to 2X higher annualized returns than classical factor libraries using 70% fewer factors, and outperforms state-of-the-art deep time-series models on real markets. Its joint factor-model optimization delivers a strong balance between predictive accuracy and strategy robustness. Our code is available at: https://github.com/microsoft/RD-Agent.

Patrick Loiseau, Simon Mauras, Minrui Xu

We study auction design in the celebrated interdependence model introduced by Milgrom and Weber [1982], where a mechanism designer allocates a good, maximizing the value of the agent who receives it, while inducing truthfulness using payments. In the lesser-studied procurement auctions, one allocates a chore, minimizing the cost incurred by the agent selected to perform it. Most of the past literature in theoretical computer science considers designing truthful mechanisms with constant approximation for the value setting, with restricted domains and monotone valuation functions. In this work, we study the general computational problems of optimizing the approximation ratio of truthful mechanism, for both value and cost, in the deterministic and randomized settings. Unlike most previous works, we remove the domain restriction and the monotonicity assumption imposed on value functions. We provide theoretical explanations for why some previously considered special cases are tractable, reducing them to classical combinatorial problems, and providing efficient algorithms and characterizations. We complement our positive results with hardness results for the general case, providing query complexity lower bounds, and proving the NP-Hardness of the general case.

Minrui Xu, Dusit Niyato, Jiawen Kang et al.

AI agents based on multimodal large language models (LLMs) are expected to revolutionize human-computer interaction and offer more personalized assistant services across various domains like healthcare, education, manufacturing, and entertainment. Deploying LLM agents in 6G networks enables users to access previously expensive AI assistant services via mobile devices democratically, thereby reducing interaction latency and better preserving user privacy. Nevertheless, the limited capacity of mobile devices constrains the effectiveness of deploying and executing local LLMs, which necessitates offloading complex tasks to global LLMs running on edge servers during long-horizon interactions. In this article, we propose a split learning system for LLM agents in 6G networks leveraging the collaboration between mobile devices and edge servers, where multiple LLMs with different roles are distributed across mobile devices and edge servers to perform user-agent interactive tasks collaboratively. In the proposed system, LLM agents are split into perception, grounding, and alignment modules, facilitating inter-module communications to meet extended user requirements on 6G network functions, including integrated sensing and communication, digital twins, and task-oriented communications. Furthermore, we introduce a novel model caching algorithm for LLMs within the proposed system to improve model utilization in context, thus reducing network costs of the collaborative mobile and edge LLM agents.

Zhicheng Yang, Zhijiang Guo, Yifan Song et al.

On-policy distillation (OPD) leverages dense teacher rewards to enhance reasoning models. However, scaling OPD to long-horizon tasks exposes a critical flaw: as the student's generated prefix inevitably diverges from the teacher's thought process, the teacher's dense reward loses local exploitability. Continuing to generate and evaluate tokens on these ``drifted'' trajectories not only degrades reward quality but also incurs massive computational waste. To address this, we introduce \textbf{Prune-OPD}, a framework that dynamically aligns training budgets with supervision quality. By continuously monitoring the local compatibility between student and teacher predictions (e.g., via top-$k$ overlap), Prune-OPD detects prefix-drift events in real time. Upon detecting severe drift, it monotonically down-weights subsequent unreliable rewards and triggers dynamic rollout truncation. This allows the training process to halt futile generation and reallocate compute strictly to reliable teacher supervision. Across diverse teacher-student combinations, Prune-OPD consistently aligns computation with supervision reliability. When prefix drift makes dense teacher rewards unreliable, it reduces training time by 37.6\%--68.0\% while preserving, and often improving, performance on challenging benchmarks (AMC, AIME, HMMT). When student-teacher compatibility remains high, it automatically preserves long-context supervision by expanding the training window. These results suggest that Prune-OPD improves OPD not by blindly shortening rollouts, but by reallocating computation toward locally exploitable teacher rewards.

Wensheng Su, Zhenni Li, Minrui Xu et al.

Deep reinforcement learning (DRL) has shown remarkable success in complex autonomous driving scenarios. However, DRL models inevitably bring high memory consumption and computation, which hinders their wide deployment in resource-limited autonomous driving devices. Structured Pruning has been recognized as a useful method to compress and accelerate DRL models, but it is still challenging to estimate the contribution of a parameter (i.e., neuron) to DRL models. In this paper, we introduce a novel dynamic structured pruning approach that gradually removes a DRL model's unimportant neurons during the training stage. Our method consists of two steps, i.e. training DRL models with a group sparse regularizer and removing unimportant neurons with a dynamic pruning threshold. To efficiently train the DRL model with a small number of important neurons, we employ a neuron-importance group sparse regularizer. In contrast to conventional regularizers, this regularizer imposes a penalty on redundant groups of neurons that do not significantly influence the output of the DRL model. Furthermore, we design a novel structured pruning strategy to dynamically determine the pruning threshold and gradually remove unimportant neurons with a binary mask. Therefore, our method can remove not only redundant groups of neurons of the DRL model but also achieve high and robust performance. Experimental results show that the proposed method is competitive with existing DRL pruning methods on discrete control environments (i.e., CartPole-v1 and LunarLander-v2) and MuJoCo continuous environments (i.e., Hopper-v3 and Walker2D-v3). Specifically, our method effectively compresses $93\%$ neurons and $96\%$ weights of the DRL model in four challenging DRL environments with slight accuracy degradation.

Minrui Xu, Dusit Niyato, Jiawen Kang et al.

Generative AI (GAI) can enhance the cognitive, reasoning, and planning capabilities of intelligent modules in the Internet of Vehicles (IoV) by synthesizing augmented datasets, completing sensor data, and making sequential decisions. In addition, the mixture of experts (MoE) can enable the distributed and collaborative execution of AI models without performance degradation between connected vehicles. In this survey, we explore the integration of MoE and GAI to enable Artificial General Intelligence in IoV, which can enable the realization of full autonomy for IoV with minimal human supervision and applicability in a wide range of mobility scenarios, including environment monitoring, traffic management, and autonomous driving. In particular, we present the fundamentals of GAI, MoE, and their interplay applications in IoV. Furthermore, we discuss the potential integration of MoE and GAI in IoV, including distributed perception and monitoring, collaborative decision-making and planning, and generative modeling and simulation. Finally, we present several potential research directions for facilitating the integration.

Jiani Fan, Minrui Xu, Ziyao Liu et al.

Artificial Intelligence-Generated Content (AIGC) refers to the paradigm of automated content generation utilizing AI models. Mobile AIGC services in the Internet of Vehicles (IoV) network have numerous advantages over traditional cloud-based AIGC services, including enhanced network efficiency, better reconfigurability, and stronger data security and privacy. Nonetheless, AIGC service provisioning frequently demands significant resources. Consequently, resource-constrained roadside units (RSUs) face challenges in maintaining a heterogeneous pool of AIGC services and addressing all user service requests without degrading overall performance. Therefore, in this paper, we propose a decentralized incentive mechanism for mobile AIGC service allocation, employing multi-agent deep reinforcement learning to find the balance between the supply of AIGC services on RSUs and user demand for services within the IoV context, optimizing user experience and minimizing transmission latency. Experimental results demonstrate that our approach achieves superior performance compared to other baseline models.

Bingkun Lai, Jiayi He, Jiawen Kang et al.

Generative Artificial Intelligence (GAI) shows remarkable productivity and creativity in Mobile Edge Networks, such as the metaverse and the Industrial Internet of Things. Federated learning is a promising technique for effectively training GAI models in mobile edge networks due to its data distribution. However, there is a notable issue with communication consumption when training large GAI models like generative diffusion models in mobile edge networks. Additionally, the substantial energy consumption associated with training diffusion-based models, along with the limited resources of edge devices and complexities of network environments, pose challenges for improving the training efficiency of GAI models. To address this challenge, we propose an on-demand quantized energy-efficient federated diffusion approach for mobile edge networks. Specifically, we first design a dynamic quantized federated diffusion training scheme considering various demands from the edge devices. Then, we study an energy efficiency problem based on specific quantization requirements. Numerical results show that our proposed method significantly reduces system energy consumption and transmitted model size compared to both baseline federated diffusion and fixed quantized federated diffusion methods while effectively maintaining reasonable quality and diversity of generated data.

Ruochen Zhou, Minrui Xu, Shiqi Chen et al.

There has been a growing interest in enhancing the mathematical problem-solving (MPS) capabilities of large language models. While the majority of research efforts concentrate on creating specialized models to solve mathematical problems, it remains unknown how learning mathematical problem-solving generalizes to help develop other reasoning abilities. In this paper, we present an empirical investigation into the generalization potential of various MPS training approaches, such as continual pretraining, instruction tuning, and rule-based reinforcement learning across various data sources, including both short and long chain-of-thought (CoT) samples. Evaluation on 5 mathematical and 8 general reasoning benchmarks show that continual pretraining on math text is able to generalize to general reasoning tasks to some extent. In constrast, instruction tuning on conventional, short MPS samples provides limited benefits and, in many cases, even impairs generalization performance. Notably, training with long CoT responses for MPS samples and incorporating rule-based reinforcement learning on MPS queries exhibit distinct behavior, significantly enhancing generalization by extending the model's reasoning processes into other domains. These results suggest that traditional approaches to learning MPS with short reasoning chains largely fail to achieve robust generalization. However, the emerging paradigm of longer reasoning chains, coupled with self-reflection, offers a promising direction for improving generalized reasoning abilities through learning from specialized domains.

Jiawen Kang, Yongju Tong, Yue Zhong et al.

The rise of 6G-enable Vehicular Metaverses is transforming the automotive industry by integrating immersive, real-time vehicular services through ultra-low latency and high bandwidth connectivity. In 6G-enable Vehicular Metaverses, vehicles are represented by Vehicle Twins (VTs), which serve as digital replicas of physical vehicles to support real-time vehicular applications such as large Artificial Intelligence (AI) model-based Augmented Reality (AR) navigation, called VT tasks. VT tasks are resource-intensive and need to be offloaded to ground Base Stations (BSs) for fast processing. However, high demand for VT tasks and limited resources of ground BSs, pose significant resource allocation challenges, particularly in densely populated urban areas like intersections. As a promising solution, Unmanned Aerial Vehicles (UAVs) act as aerial edge servers to dynamically assist ground BSs in handling VT tasks, relieving resource pressure on ground BSs. However, due to high mobility of UAVs, there exists information asymmetry regarding VT task demands between UAVs and ground BSs, resulting in inefficient resource allocation of UAVs. To address these challenges, we propose a learning-based Modified Second-Bid (MSB) auction mechanism to optimize resource allocation between ground BSs and UAVs by accounting for VT task latency and accuracy. Moreover, we design a diffusion-based reinforcement learning algorithm to optimize the price scaling factor, maximizing the total surplus of resource providers and minimizing VT task latency. Finally, simulation results demonstrate that the proposed diffusion-based MSB auction outperforms traditional baselines, providing better resource distribution and enhanced service quality for vehicular users.

Rui Pan, Dylan Zhang, Hanning Zhang et al.

Bilevel optimization has shown its utility across various machine learning settings, yet most algorithms in practice require second-order information, making it challenging to scale them up. Only recently, a paradigm of first-order algorithms has emerged in the theoretical literature, capable of effectively addressing bilevel optimization problems. Nevertheless, the practical efficiency of this paradigm remains unverified, particularly in the context of large language models (LLMs). This paper introduces the first scalable instantiation of this paradigm called ScaleBiO, focusing on bilevel optimization for large-scale LLM data reweighting. By combining with a recently proposed memory-efficient training technique called LISA, our novel algorithm allows the paradigm to scale to $\sim$30B-sized LLMs on $8\times$H100 GPUs, marking the first successful application of bilevel optimization under practical scenarios for large-sized LLMs. Empirically, extensive experiments on data reweighting verify the effectiveness of ScaleBiO for different-scaled models, including Llama-3-8B, Gemma-2-9B, Qwen-2-7B, and Qwen-2.5-32B, where bilevel optimization succeeds in instruction-following and math reasoning tasks, outperforming several popular baselines, including uniform sampling, influence-aware data filtering, and reference-model-based sampling methods. Theoretically, ScaleBiO ensures the optimality of the learned data weights, along with a convergence guarantee matching the conventional first-order bilevel optimization paradigm on smooth and strongly convex objectives.

Yongju Tong, Jiawen Kang, Junlong Chen et al.

Air-ground integrated networks can relieve communication pressure on ground transportation networks and provide 6G-enabled vehicular Metaverses services offloading in remote areas with sparse RoadSide Units (RSUs) coverage and downtown areas where users have a high demand for vehicular services. Vehicle Twins (VTs) are the digital twins of physical vehicles to enable more immersive and realistic vehicular services, which can be offloaded and updated on RSU, to manage and provide vehicular Metaverses services to passengers and drivers. The high mobility of vehicles and the limited coverage of RSU signals necessitate VT migration to ensure service continuity when vehicles leave the signal coverage of RSUs. However, uneven VT task migration might overload some RSUs, which might result in increased service latency, and thus impactive immersive experiences for users. In this paper, we propose a dynamic Unmanned Aerial Vehicle (UAV)-assisted VT migration framework in air-ground integrated networks, where UAVs act as aerial edge servers to assist ground RSUs during VT task offloading. In this framework, we propose a diffusion-based Reinforcement Learning (RL) algorithm, which can efficiently make immersive VT migration decisions in UAV-assisted vehicular networks. To balance the workload of RSUs and improve VT migration quality, we design a novel dynamic path planning algorithm based on a heuristic search strategy for UAVs. Simulation results show that the diffusion-based RL algorithm with UAV-assisted performs better than other baseline schemes.

Yongju Tong, Junlong Chen, Minrui Xu et al.

Vehicular Metaverses are developed to enhance the modern automotive industry with an immersive and safe experience among connected vehicles and roadside infrastructures, e.g., RoadSide Units (RSUs). For seamless synchronization with virtual spaces, Vehicle Twins (VTs) are constructed as digital representations of physical entities. However, resource-intensive VTs updating and high mobility of vehicles require intensive computation, communication, and storage resources, especially for their migration among RSUs with limited coverages. To address these issues, we propose an attribute-aware auction-based mechanism to optimize resource allocation during VTs migration by considering both price and non-monetary attributes, e.g., location and reputation. In this mechanism, we propose a two-stage matching for vehicular users and Metaverse service providers in multi-attribute resource markets. First, the resource attributes matching algorithm obtains the resource attributes perfect matching, namely, buyers and sellers can participate in a double Dutch auction (DDA). Then, we train a DDA auctioneer using a generative pre-trained transformer (GPT)-based deep reinforcement learning (DRL) algorithm to adjust the auction clocks efficiently during the auction process. We compare the performance of social welfare and auction information exchange costs with state-of-the-art baselines under different settings. Simulation results show that our proposed GPT-based DRL auction schemes have better performance than others.

Jiawen Kang, Yue Zhong, Minrui Xu et al.

The synergy between Unmanned Aerial Vehicles (UAVs) and metaverses is giving rise to an emerging paradigm named UAV metaverses, which create a unified ecosystem that blends physical and virtual spaces, transforming drone interaction and virtual exploration. UAV Twins (UTs), as the digital twins of UAVs that revolutionize UAV applications by making them more immersive, realistic, and informative, are deployed and updated on ground base stations, e.g., RoadSide Units (RSUs), to offer metaverse services for UAV Metaverse Users (UMUs). Due to the dynamic mobility of UAVs and limited communication coverages of RSUs, it is essential to perform real-time UT migration to ensure seamless immersive experiences for UMUs. However, selecting appropriate RSUs and optimizing the required bandwidth is challenging for achieving reliable and efficient UT migration. To address the challenges, we propose a tiny machine learning-based Stackelberg game framework based on pruning techniques for efficient UT migration in UAV metaverses. Specifically, we formulate a multi-leader multi-follower Stackelberg model considering a new immersion metric of UMUs in the utilities of UAVs. Then, we design a Tiny Multi-Agent Deep Reinforcement Learning (Tiny MADRL) algorithm to obtain the tiny networks representing the optimal game solution. Specifically, the actor-critic network leverages the pruning techniques to reduce the number of network parameters and achieve model size and computation reduction, allowing for efficient implementation of Tiny MADRL. Numerical results demonstrate that our proposed schemes have better performance than traditional schemes.