arXiv subjects

Xian Wang

Publications and source records attributed to Xian Wang.

At least 19 recordsLinked to original sources

arXiv2026

DP-LENS: A Density-Aware Polyfocal Lens with Topology-Driven Auto-Routing for Occlusion Management in Immersive 3D Analytics

Immersive environments, e.g., virtual reality (VR), offer a unique approach to exploring complex 3D datasets, where data is often heavily occluded and exploration incurs a high cognitive load. We propose DP-LENS, a density-aware polyfocal fisheye lens equipped with topology-driven auto-routing. While preserving peripheral context through geometric deformation and 3D perspective techniques, it enables users to explore 3D data with a lower cognitive load. To facilitate hands-free macro-navigation, we integrate a Large Language Model (LLM) to serve as a supplementary voice-based target selection tool that initiates the auto-routing algorithm. Two user studies with 34 participants investigate the potential benefits of this system. Our first study (N=18) compared the manual DP-LENS against two industry-standard baselines (i.e., World-in-Miniature and volumetric slicing) in heavily occluded 3D datasets. The results show that DP-LENS significantly reduced cognitive load, decreased completion time, and improved user preference. The second study (N=16) compared the topology-driven auto-routing system (initiated via voice commands) with a fully manual DP-LENS. The results show that the auto-routing system improved task efficiency, further reduced cognitive load, and garnered higher user preference. Furthermore, the auto-routing partially decoupled exploration efficiency from the physical dimensions of the data and mitigated physical fatigue to some extent. Based on the findings, we proposed design implications to inform the development of more spatially scalable and low-fatigue interactions for future 3D visual analytics systems.

cs.HC↗

arXiv2026

A GPGPU-Oriented Full Phase-Space Parallel Unified Gas-Kinetic Scheme with Velocity-Block Pipelining

The deterministic unified gas-kinetic scheme (UGKS) provides a multiscale framework for nonequilibrium gas dynamics, but its high-dimensional phase-space discretization leads to severe memory pressure and communication overhead, especially on large unstructured meshes. This paper presents a GPGPU-oriented UGKS with velocity-block pipelining and full phase-space MPI decomposition. In the proposed formulation, the discrete velocity space is partitioned into fixed-size velocity blocks for accelerator execution, while MPI ranks are organized into coupled physical-space and velocity-space communicators. As a result, each rank stores and advances only a local physical subdomain together with a contiguous subset of velocity blocks, and macroscopic moments are recovered through lightweight reductions over the velocity-space communicator. To improve concurrency and reduce exposed communication cost, a triple-buffered pipeline is further developed to overlap microscopic reconstruction, physical-halo exchange, nonequilibrium flux evaluation, and the first-stage distribution update during the local velocity-block sweep. The implementation targets SIMT-based GPGPU accelerators through a portable device-runtime abstraction. Numerical experiments demonstrate that the $P_v=8$ configuration achieves a $33.4$--$35.4\times$ strong-scaling speedup on 64 nodes, while an Orion-like capsule simulation reaches approximately $1.33\times10^{11}$ phase-space degrees of freedom on 4096 GPGPU accelerators. These results indicate that the proposed method preserves the original UGKS flux construction and two-stage time discretization, while substantially reducing microscopic storage per rank and improving the scalability of large unstructured phase-space simulations.

physics.comp-ph↗

arXiv2026

The Third Challenge on Image Denoising at NTIRE 2026: Methods and Results

This paper reports on the NTIRE 2026 Challenge on Image Denoising, specifically focusing on the high-noise regime ($\sigma = 50$). The competition investigates advanced neural architectures designed to restore high-fidelity details from images corrupted by additive white Gaussian noise (AWGN). Unlike constrained benchmarks, this track emphasizes peak quantitative performance, measured by Peak Signal-to-Noise Ratio (PSNR), without limitations on parameter count or computational overhead. By synthesizing contributions from 20 finalist teams out of 116 registrants, this report benchmarks the latest technical innovations and provides a comprehensive snapshot of the current state-of-the-art in unconstrained image restoration.

cs.CV↗

arXiv2026

Masked Generative Transformer Is What You Need for Image Editing

Diffusion models dominate image editing, yet their global denoising mechanism entangles edited regions with surrounding context, causing modifications to propagate into areas that should remain intact. We propose a fundamentally different approach by leveraging Masked Generative Transformers (MGTs), whose localized token-prediction paradigm naturally confines changes to intended regions. We present EditMGT, an MGT-based editing framework that is the first of its kind. Our approach employs multi-layer attention consolidation to aggregate cross-attention maps into precise edit localization signals, and region-hold sampling to explicitly prevent token flipping in non-target areas. To support training, we construct CrispEdit-2M, a 2M-sample high-resolution (>1024) editing dataset spanning seven categories. With only 960M parameters, EditMGT achieves state-of-the-art image similarity on multiple benchmarks while delivering 6x faster editing, demonstrating that MGTs offer a compelling alternative to diffusion-based editing.

cs.CV↗

arXiv2026

Conflict Resolution Strategies for Co-manipulation of Virtual Objects Under Non-disjoint Conditions

Virtual Reality (VR) co-manipulation enables multiple users to collaboratively interact with shared virtual objects. However, existing research treats objects as monolithic entities, overlooking scenarios where users need to manipulate different sub-components simultaneously. This work addresses conflict resolution when users select overlapping vertices (non-disjoint sets) during co-manipulation. We present a comprehensive framework comprising preventive strategies (Object-level and Action-level Restrictions) and reactive strategies (computational conflict resolution). Through two user studies with 76 participants (38 pairs), we evaluated these approaches in collaborative wireframe editing tasks. Study 1 identified Averaging as the optimal computational method, balancing task efficiency with user experience. Study 2 highlighted that Action-level Restriction, which permits overlapping selections but restricts concurrent identical operations, achieved better performance compared to exclusive object locking. Reactive strategies using averaging provided smooth collaboration for experienced users, while second-user priority enabled quick corrections. Our findings indicate that optimal strategy selection depends on task requirements, user expertise, and collaboration patterns. Based on the findings, we provide design implications for developing VR collaboration systems that support flexible sub-components manipulation while maintaining collaborative awareness and minimizing conflicts.

cs.HC↗

arXiv2026

AgentCollab: A Self-Evaluation-Driven Collaboration Paradigm for Efficient LLM Agents

Autonomous agents powered by large language models (LLMs) perform complex tasks through long-horizon reasoning and tool interaction, where a fundamental trade-off arises between execution efficiency and reasoning robustness. Models at different capability-cost levels offer complementary advantages: lower-cost models enable fast execution but may struggle on difficult reasoning segments, while stronger models provide more robust reasoning at higher computational cost. We present AgentCollab, a self-driven collaborative inference framework that dynamically coordinates models with different reasoning capacities during agent execution. Instead of relying on external routing modules, the framework uses the agent's own self-reflection signal to determine whether the current reasoning trajectory is making meaningful progress, and escalates control to a stronger reasoning tier only when necessary. To further stabilize long-horizon execution, we introduce a difficulty-aware cumulative escalation strategy that allocates additional reasoning budget based on recent failure signals. In our experiments, we instantiate this framework using a two-level small-large model setting. Experiments on diverse multi-step agent benchmarks show that AgentCollab consistently improves the accuracy-efficiency Pareto frontier of LLM agents.

cs.CL↗

arXiv2026

MAVEN: A Meta-Reinforcement Learning Framework for Varying-Dynamics Expertise in Agile Quadrotor Maneuvers

Reinforcement learning (RL) has emerged as a powerful paradigm for achieving online agile navigation with quadrotors. Despite this success, policies trained via standard RL typically fail to generalize across significant dynamic variations, exhibiting a critical lack of adaptability. This work introduces MAVEN, a meta-RL framework that enables a single policy to achieve robust end-to-end navigation across a wide range of quadrotor dynamics. Our approach features a novel predictive context encoder, which learns to infer a latent representation of the system dynamics from interaction history. We demonstrate our method in agile waypoint traversal tasks under two challenging scenarios: large variations in quadrotor mass and severe single-rotor thrust loss. We leverage a GPU-vectorized simulator to distribute tasks across thousands of parallel environments, overcoming the long training times of meta-RL to converge in less than an hour. Through extensive experiments in both simulation and the real world, we validate that MAVEN achieves superior adaptation and agility. The policy successfully executes zero-shot sim-to-real transfer, demonstrating robust online adaptation by performing high-speed maneuvers despite mass variations of up to 66.7% and single-rotor thrust losses as severe as 70%.

cs.RO↗

arXiv2026

Non-urgent Messages Do Not Jump into My Headset Suddenly! Adaptive Notification Design in Mixed Reality

Mixed reality (MR) notification systems currently display all messages in fixed central locations regardless of urgency, leading to unnecessary interruptions and cognitive overload. Drawing from previous MR/Virtual Reality (VR) notification design work and calm technology principles, we developed an adaptive notification system that adjusts spatial placement based on urgency levels: non-urgent notifications appear as peripheral icons accessible via head movement, moderately urgent messages anchor to the user's hand, and very urgent notifications transition progressively from peripheral to central view. Through a within-subjects study (N=18), we evaluated our adaptive system against the default centralised approach. Results demonstrate that the adaptive system significantly reduces mental workload (p=0.041), temporal workload (p=0.008), and frustration (p=0.004) while maintaining comparable notification awareness. Logistic regression analysis reveals that users prefer the adaptive system even with classification errors, provided the combined misclassification rate (disruptiveness + omission errors) remains below a determinable threshold. Our findings establish the first empirical evidence that urgency-based spatial notification distribution effectively addresses core MR usability challenges, offering practical design guidelines for immersive notification systems that balance user attention management with information accessibility.

cs.HC↗

arXiv2025

Towards Efficient Agents: A Co-Design of Inference Architecture and System

The rapid development of large language model (LLM)-based agents has unlocked new possibilities for autonomous multi-turn reasoning and tool-augmented decision-making. However, their real-world deployment is hindered by severe inefficiencies that arise not from isolated model inference, but from the systemic latency accumulated across reasoning loops, context growth, and heterogeneous tool interactions. This paper presents AgentInfer, a unified framework for end-to-end agent acceleration that bridges inference optimization and architectural design. We decompose the problem into four synergistic components: AgentCollab, a hierarchical dual-model reasoning framework that balances large- and small-model usage through dynamic role assignment; AgentSched, a cache-aware hybrid scheduler that minimizes latency under heterogeneous request patterns; AgentSAM, a suffix-automaton-based speculative decoding method that reuses multi-session semantic memory to achieve low-overhead inference acceleration; and AgentCompress, a semantic compression mechanism that asynchronously distills and reorganizes agent memory without disrupting ongoing reasoning. Together, these modules form a Self-Evolution Engine capable of sustaining efficiency and cognitive stability throughout long-horizon reasoning tasks. Experiments on the BrowseComp-zh and DeepDiver benchmarks demonstrate that through the synergistic collaboration of these methods, AgentInfer reduces ineffective token consumption by over 50%, achieving an overall 1.8-2.5 times speedup with preserved accuracy. These results underscore that optimizing for agentic task completion-rather than merely per-token throughput-is the key to building scalable, efficient, and self-improving intelligent systems.

cs.CL↗

arXiv2025

EditMGT: Unleashing Potentials of Masked Generative Transformers in Image Editing

Recent advances in diffusion models (DMs) have achieved exceptional visual quality in image editing tasks. However, the global denoising dynamics of DMs inherently conflate local editing targets with the full-image context, leading to unintended modifications in non-target regions. In this paper, we shift our attention beyond DMs and turn to Masked Generative Transformers (MGTs) as an alternative approach to tackle this challenge. By predicting multiple masked tokens rather than holistic refinement, MGTs exhibit a localized decoding paradigm that endows them with the inherent capacity to explicitly preserve non-relevant regions during the editing process. Building upon this insight, we introduce the first MGT-based image editing framework, termed EditMGT. We first demonstrate that MGT's cross-attention maps provide informative localization signals for localizing edit-relevant regions and devise a multi-layer attention consolidation scheme that refines these maps to achieve fine-grained and precise localization. On top of these adaptive localization results, we introduce region-hold sampling, which restricts token flipping within low-attention areas to suppress spurious edits, thereby confining modifications to the intended target regions and preserving the integrity of surrounding non-target areas. To train EditMGT, we construct CrispEdit-2M, a high-resolution dataset spanning seven diverse editing categories. Without introducing additional parameters, we adapt a pre-trained text-to-image MGT into an image editing model through attention injection. Extensive experiments across four standard benchmarks demonstrate that, with fewer than 1B parameters, our model achieves similarity performance while enabling 6 times faster editing. Moreover, it delivers comparable or superior editing quality, with improvements of 3.6% and 17.6% on style change and style transfer tasks, respectively.

cs.CV↗

arXiv2025

Towards Blind Lens Aberration Correction via Large LensLib Pre-training and Discrete Degradation Priors

Emerging deep-learning-based lens library pre-training (LensLib-PT) pipeline offers a new avenue for blind lens aberration correction by training a universal neural network, demonstrating strong capability in handling diverse unknown optical degradations. This work proposes FoundCAC, a universal foundational framework that resolves two challenges hindering the generalization of existing pipelines: the difficulty of scaling training data and the absence of prior guidance characterizing optical degradation. To improve data scalability, we expand the design specifications to increase degradation diversity and construct AODLibpro, a large-scale lens library using stratified sampling over spatial-variation patterns and degradation severity. In terms of model design, to leverage Point Spread Functions (PSFs) as guidance while maintaining the blind paradigm, we propose a multi-stage vector-quantized representation learning scheme. This paradigm is specifically designed to construct a Latent PSF Representation (LPR), explicitly encoding complex continuous PSFs into a discrete degradation prior to regularize the highly ill-posed restoration process. Through a simple yet effective codebook-freezing strategy, our framework leverages the discrete prior to elevate full-shot restoration performance and unlock highly efficient few-shot adaptation for unseen lenses. Experiments on synthetic LensLib, real-design simulations, and real-captured lenses show that our framework achieves state-of-the-art zero-shot performance under complementary evaluation protocols, while enabling highly efficient few-shot adaptation for specific lenses. The source code and datasets will be made publicly available at https://github.com/zju-jiangqi/FoundCAC.

eess.IV↗

arXiv2025

Equivariant Atomic and Lattice Modeling Using Geometric Deep Learning for Crystal Structure Optimization

Structure optimization, which yields the relaxed structure (minimum-energy state), is essential for reliable materials property calculations, yet traditional ab initio approaches such as density-functional theory (DFT) are computationally intensive. Machine learning (ML) has emerged to alleviate this bottleneck but suffers from two major limitations: (i) existing models operate mainly on atoms, leaving lattice vectors implicit despite their critical role in structural optimization; and (ii) they often rely on multi-stage, non-end-to-end workflows that are prone to error accumulation. Here, we present E3Relax, an end-to-end equivariant graph neural network that maps an unrelaxed crystal directly to its relaxed structure. E3Relax promotes both atoms and lattice vectors to graph nodes endowed with dual scalar-vector features, enabling unified and symmetry-preserving modeling of atomic displacements and lattice deformations. A layer-wise supervision strategy forces every network depth to make a physically meaningful refinement, mimicking the incremental convergence of DFT while preserving a fully end-to-end pipeline. We evaluate E3Relax on four benchmark datasets and demonstrate that it achieves remarkable accuracy and efficiency. Through DFT validations, we show that the structures predicted by E3Relax are energetically favorable, making them suitable as high-quality initial configurations to accelerate DFT calculations.

cond-mat.mtrl-sci↗

arXiv2025

When Generative Artificial Intelligence meets Extended Reality: A Systematic Review

With the continuous advancement of technology, the application of generative artificial intelligence (AI) in various fields is gradually demonstrating great potential, particularly when combined with Extended Reality (XR), creating unprecedented possibilities. This survey article systematically reviews the applications of generative AI in XR, covering as much relevant literature as possible from 2023 to 2025. The application areas of generative AI in XR and its key technology implementations are summarised through PRISMA screening and analysis of the final 26 articles. The survey highlights existing articles from the last three years related to how XR utilises generative AI, providing insights into current trends and research gaps. We also explore potential opportunities for future research to further empower XR through generative AI, providing guidance and information for future generative XR research.

cs.HC↗

arXiv2025

MIRAGE: Multimodal Intention Recognition and Admittance-Guided Enhancement in VR-based Multi-object Teleoperation

Effective human-robot interaction (HRI) in multi-object teleoperation tasks faces significant challenges due to perceptual ambiguities in virtual reality (VR) environments and the limitations of single-modality intention recognition. This paper proposes a shared control framework that combines a virtual admittance (VA) model with a Multimodal-CNN-based Human Intention Perception Network (MMIPN) to enhance teleoperation performance and user experience. The VA model employs artificial potential fields to guide operators toward target objects by adjusting admittance force and optimizing motion trajectories. MMIPN processes multimodal inputs, including gaze movement, robot motions, and environmental context, to estimate human grasping intentions, helping to overcome depth perception challenges in VR. Our user study evaluated four conditions across two factors, and the results showed that MMIPN significantly improved grasp success rates, while the VA model enhanced movement efficiency by reducing path lengths. Gaze data emerged as the most crucial input modality. These findings demonstrate the effectiveness of combining multimodal cues with implicit guidance in VR-based teleoperation, providing a robust solution for multi-object grasping tasks and enabling more natural interactions across various applications in the future.

cs.RO↗

arXiv2025

MetaRoundWorm: A Virtual Reality Escape Room Game for Learning the Lifecycle and Immune Response to Parasitic Infections

Promoting public health is challenging owing to its abstract nature, and individuals may be apprehensive about confronting it. Recently, there has been an increasing interest in using the metaverse and gamification as novel educational techniques to improve learning experiences related to the immune system. Thus, we present MetaRoundWorm, an immersive virtual reality (VR) escape room game designed to enhance the understanding of parasitic infections and host immune responses through interactive, gamified learning. The application simulates the lifecycle of Ascaris lumbricoides and corresponding immunological mechanisms across anatomically accurate environments within the human body. Integrating serious game mechanics with embodied learning principles, MetaRoundWorm offers players a task-driven experience combining exploration, puzzle-solving, and immune system simulation. To evaluate the educational efficacy and user engagement, we conducted a controlled study comparing MetaRoundWorm against a traditional approach, i.e., interactive slides. Results indicate that MetaRoundWorm significantly improves immediate learning outcomes, cognitive engagement, and emotional experience, while maintaining knowledge retention over time. Our findings suggest that immersive VR gamification holds promise as an effective pedagogical tool for communicating complex biomedical concepts and advancing digital health education.

cs.HC↗

arXiv2025

PersoNo: Personalised Notification Urgency Classifier in Mixed Reality

Mixed Reality (MR) is increasingly integrated into daily life, providing enhanced capabilities across various domains. However, users face growing notification streams that disrupt their immersive experience. We present PersoNo, a personalised notification urgency classifier for MR that intelligently classifies notifications based on individual user preferences. Through a user study (N=18), we created the first MR notification dataset containing both self-labelled and interaction-based data across activities with varying cognitive demands. Our thematic analysis revealed that, unlike in mobiles, the activity context is equally important as the content and the sender in determining notification urgency in MR. Leveraging these insights, we developed PersoNo using large language models that analyse users replying behaviour patterns. Our multi-agent approach achieved 81.5% accuracy and significantly reduced false negative rates (0.381) compared to baseline models. PersoNo has the potential not only to reduce unnecessary interruptions but also to offer users understanding and control of the system, adhering to Human-Centered Artificial Intelligence design principles.

cs.HC↗

arXiv2025

FLARE: Agile Flights for Quadrotor Cable-Suspended Payload System via Reinforcement Learning

Agile flight for the quadrotor cable-suspended payload system is a formidable challenge due to its underactuated, highly nonlinear, and hybrid dynamics. Traditional optimization-based methods often struggle with high computational costs and the complexities of cable mode transitions, limiting their real-time applicability and maneuverability exploitation. In this letter, we present FLARE, a reinforcement learning (RL) framework that directly learns agile navigation policy from high-fidelity simulation. Our method is validated across three designed challenging scenarios, notably outperforming a state-of-the-art optimization-based approach by a 3x speedup during gate traversal maneuvers. Furthermore, the learned policies achieve successful zero-shot sim-to-real transfer, demonstrating remarkable agility and safety in real-world experiments, running in real time on an onboard computer.

cs.RO↗

arXiv2025

Seeing is Believing? Mitigating OCR Hallucinations in Multimodal Large Language Models

Recent advancements in multimodal large language models have enhanced document understanding by integrating textual and visual information. However, existing models exhibit incompleteness within their paradigm in real-world scenarios, particularly under visual degradation. In such conditions, the current response paradigm often fails to adequately perceive visual degradation and ambiguity, leading to overreliance on linguistic priors or misaligned visual-textual reasoning. This difficulty in recognizing uncertainty frequently results in the generation of hallucinatory content, especially when a precise answer is not feasible. To better demonstrate and analyze this phenomenon and problem, we propose KIE-HVQA, the first benchmark dedicated to evaluating OCR hallucination in degraded document understanding. This dataset includes test samples spanning identity cards and invoices, with simulated real-world degradations for OCR reliability. This setup allows for evaluating models' capacity, under degraded input, to distinguish reliable visual information and answer accordingly, thereby highlighting the challenge of avoiding hallucination on uncertain data. To achieve vision-faithful reasoning and thereby avoid the aforementioned issues, we further introduce a GRPO-based framework featuring a novel reward mechanism. By incorporating a self-awareness of visual uncertainty and an analysis method that initiates refusal to answer to increase task difficulty within our supervised fine-tuning and reinforcement learning framework, we successfully mitigated hallucinations in ambiguous regions. Experiments on Qwen2.5-VL demonstrate that our 7B-parameter model achieves a 22\% absolute improvement in hallucination-free accuracy over GPT-4o on KIE-HVQA and there is no significant performance drop in standard tasks, highlighting both effectiveness and robustness.

cs.CV↗