arXiv subjects

Yan Zhang

Publications and source records attributed to Yan Zhang.

At least 73 records · Page 4Linked to original sources

arXiv2026

PAIWorld: A 3D-Consistent World Foundation Model for Robotic Manipulation

World foundation models (WFMs) are powerful simulators, yet they predominantly operate in a single-view setting and lack the multi-view 3D consistency required for robotic manipulation. While robotic systems rely on multiple cameras (egocentric, eye-to-hand, and wrist-mounted) for policy learning, current multi-view world models simply concatenate view tokens without explicit geometric reasoning. This causes cross-view object drift, depth inconsistency, and texture misalignment. We trace these failures to two deficiencies: the absence of an explicit inter-view communication mechanism and the lack of a 3D geometric prior. We argue that resolving both simultaneously is necessary and sufficient. To address this, we present PAIWorld, a framework that augments diffusion-transformer world models via three core components: (1) Geometry-Aware Cross-View Attention blocks that establish an explicit pathway across views, (2) Geometric Rotary Position Embedding that encodes camera ray directions and extrinsic poses into the attention mechanism, and (3) Latent 3D-REPA, which distills 3D-aware features from frozen 3D foundation models to ensure 3D consistency. Built upon a DiT-based world foundation model, PAIWorld achieves state-of-the-art multi-view 3D consistency on robotic manipulation benchmarks, ranking 1st on the WorldArena leaderboard and 2nd on the AgiBot-Challenge2026 leaderboard, while enabling downstream applications such as model-based planning, world action models, and multi-view policy post-training.

cs.RO↗

arXiv2026

Autonomous Generation of Metamaterial Databases Based on Multimodal Agents

Artificial intelligence (AI) is revolutionizing material research and discovery. However, its development in metamaterials is bottlenecked by a shortage of high-quality and executable structure-response databases, which are locked within scientific literatures as a mixture of text and images. Converting the rapidly growing body of scientific literatures into executable and reusable databases for machine-driven discovery is still a fundamental challenge. Here, we propose MetaDataGenAgent, a multimodal multi-agent framework that autonomously converts unstructured scientific literatures directly into metamaterial structure-response databases. MetaDataGenAgent establishes a complete literature-to-simulation pipeline through the coordinated operation of specialized agents for multimodal parameter extraction, physics-guided validation, topology-aware structural analysis, and solver-executable encoding. The framework introduces a closed-loop plan-execute-reflect mechanism that enables dynamic task decomposition, iterative validation, and feedback-driven model construction. Experimental results validate that MetaDataGenAgent can generate high-fidelity structure-response data for representative meta-atoms, which are further used to realize diverse electromagnetic functions, including far-field beam deflection, near-field holographic imaging and topologically protected surface-wave transport. By establishing an autonomous route from scientific literatures to AI-ready databases, the framework provides a general and efficient strategy that could be extended to a broad range of data-scarce scientific domains, including photonics, materials science, chemistry, computational science, and scientific automation.

physics.optics↗

arXiv2026

Physics-Informed Eikonal Caging for Whole-Arm Manipulation Planning

Planning contact-rich whole-arm manipulation is challenging because interactions that involve extended robot geometry give rise to complex contact dynamics that are difficult to model accurately. This creates a need for planning principles that do not rely heavily on precise contact models. Caging offers one such geometric notion of robustness to modeling inaccuracy by restricting object escape through geometrically enclosing the object. However, existing caging formulations are difficult to incorporate into continuous optimization-based manipulation planning. We reformulate caging as a minimum-time escape problem in which the object seeks to leave an enclosing robot geometry in the shortest time. This yields a continuous escape-time field that measures the robot's enclosure quality and we show it satisfies an eikonal equation. We therefore can approximate this field using a physics-informed neural network, producing a smooth differentiable representation that can be embedded directly into manipulation planning. The resulting objective supports whole-arm manipulation planning to favor robot configurations resisting object escape. This improves the manipulation robustness to contact model mismatch, thus enabling planning with simplified contact models, including quasi-dynamic approximations and simplified object geometry. Across simulation and real-world experiments, we show improved robustness to disturbances and contact-model mismatch relative to baselines. These results suggest that geometric enclosure can serve as a practical robustness primitive for whole-arm manipulation. A supplementary video, which includes an intuitive overview of our method and experiment video results, is available on our project webpage.

cs.RO↗

arXiv2026

Multi-Modal Hyper-Graph Fusion for Low-Light Crowd Counting

Crowd counting is a fundamental task in computer vision. However, crowd counting in low-light environments remains largely underexplored, despite its practical importance in the real world. Existing methods mainly focus on well-lit scenes or rely on single-modality Red-Green-Blue (RGB) representations, which often become unreliable under extreme darkness and complex non-uniform illumination. To handle this problem, we construct three new low-light crowd counting benchmarks, which consist of two synthetic datasets, SHA\_Dark and SHB\_Dark, and a real-world benchmark LC-Crowd (Low-light Crowd Dataset). Inspired by Retinex-based physical modeling, we introduce depth and Canny edge cues as complementary geometric and structural priors to enhance the intrinsic reflectance representation under low-light conditions. We propose a Multi-Modal Hyper-Graph Fusion module, which formulates RGB appearance, depth geometry, and edge structure cues as nodes in a unified hyper-graph and explicitly captures their high-order complementary relationships via dynamic hyperedge construction and message passing. Furthermore, to adaptively allocate computation in dense prediction, we propose a Deformable Rectangular Sparse Attention (DRSA) module, which concentrates computation on informative regions through anchor-aware estimation and adaptive rectangular window modeling. Based on these designs, we develop a unified Low-Light Counting Network (LCNet) for robust low-light crowd counting. Extensive experiments on three benchmarks demonstrate that the proposed method achieves the best overall performance against existing state-of-the-art (SOTA) methods. The code is in the supplementary material. The datasets will be made public upon acceptance.

cs.CV↗

arXiv2026

Hierarchical Fine-Grained Aerial Object Detection

Fine-grained aerial object detection, driven by the intrinsic granularity of real-world object categories, is crucial for advanced scene understanding in remote sensing. Existing methods largely inherit the paradigm of coarse-grained object detection, relying solely on single-label supervision and thus struggling to distinguish model-level categories with subtle structural differences. However, for each specific model (e.g., Boeing 787), structured prior knowledge such as attributes and hierarchies offers discriminative semantics across multiple granularities. Motivated by this, we present ExpertDet, a scheme that incorporates expert-informed cues to enhance fine-grained aerial object detection. Specifically, we design Vision-aware Masked Attribute Modeling (VMAM), which aligns attribute semantics with visual structures by reconstructing randomly masked attributes from visual cues, enabling the detector to capture subtle structural distinctions. We further propose Hierarchical Visual Instance Promotion (HierVIP), which builds a visual prototype tree based on hierarchical relations and imposes taxonomy-aware constraints to preserve cross-level semantic continuity while enhancing category discrimination. Moreover, we curate a new fine-grained object detection benchmark for Precise recognition of model-specific Ships and Planes from aerial imagery, PSP, covering 106 ship classes and 30 airplane models, respectively, featuring the most extensive collection of model-specific categories among existing aerial object detection datasets to date. We benchmark state-of-the-art object detection algorithms on the PSP benchmark. Extensive evaluation demonstrates that ExpertDet consistently outperforms other fine-grained competitors across hierarchy levels. The dataset, benchmark, and code are available at https://nnnnerd.github.io/PSP-Benchmark/.

cs.CV↗

arXiv2026

Towards Ubiquitous 6G Computing and Networking Convergence: Architecture and Mechanism for Cross-Domain Resource Coordination

The 6G network will support six major application scenarios, such as immersive communication, integrated AI and communication, and integrated sensing and communication. Many scenarios necessitate significant computational support. Moreover, user demands are becoming increasingly segmented, diverse, and personalized. Traditional network slicing alone is insufficient to meet the heterogeneous computing and networking demands of emerging service scenarios. Mobile computing network convergence (CNC) introduces a fundamentally different paradigm from the conventional cloud computing plus communication network model by deeply embedding computing resources into the mobile network infrastructure and enabling integrated computing-network services tailored to diverse user demands. In this article, we investigate orchestration architectures and mechanisms for CNC in 6G mobile networks. We begin by reviewing the evolution of CNC from a mobile network perspective and surveying existing studies, which we categorize according to mobile network architectures. Building on these insights, we propose a hierarchical, cross-domain coordination architecture and an orchestration mechanism based on hierarchical multi-agent reinforcement learning. Performance evaluations demonstrate that the proposed architecture and mechanism significantly reduce system energy consumption while enhancing task satisfaction rate. Finally, we discuss open challenges and future research directions.

cs.NI↗

arXiv2026

Fast computation and theoretical guarantees for the NPMLE in exponential family mixtures

This work makes two advances in the study of the (approximate) nonparametric maximum likelihood estimator (NPMLE) for exponential family mixture models. First, we develop a data-compression strategy that reduces the cost of repeated likelihood evaluations in NPMLE computation to polylogarithmic order in the sample size. Second, we show that, for a broad class of approximate NPMLEs, the resulting marginal density estimator attains an almost parametric rate of convergence.

math.ST↗

arXiv2026

Radiology-Report Semantic Modelling and Host-Response Laboratory Biomarkers for Multimodal Survival Prediction in Lung Cancer

TNM staging is essential for lung cancer management, but patients within the same anatomic stage often show heterogeneous survival outcomes. We developed a multimodal adaptive risk score (AMRS) that integrates radiology-report semantics with routinely available clinical laboratory biomarkers. In a retrospective two-center cohort, 1129 patients diagnosed between December 2017 and February 2026 were screened; 574 patients were included after exclusion for short follow-up or missing imaging reports and were split into training (n = 459) and test (n = 115) cohorts. Radiology reports were encoded with a domain-adapted MC-BERT branch to capture imaging-derived semantic information, while clinical and laboratory variables were modeled after Mahalanobis-distance-based imputation using random survival forests. Weighted risk fusion generated the final patient-level score. AMRS achieved C-index values of 0.920 in training and 0.849 in testing, and separated survival trajectories across clinical subgroups and TNM-related strata. SHAP analysis identified hematologic, inflammatory, coagulation, nutritional, tumor-marker, organ-function, and age-related contributors. AMRS may complement TNM staging in imaging-centered oncology workflows, but prospective validation, calibration, ablation testing, and clinical-utility assessment are required before deployment.

physics.med-ph↗

arXiv2026

Andreev Reflection to Probe Momentum-Dependent Spin Polarization in Altermagnet CrSb

Altermagnetic materials have recently emerged as promising candidates for next-generation spintronic applications, characterized by the k-dependent spin-splitted band structure and a simultaneous zero-net-magnetization. Among them, altermagnetic candidate CrSb has attracted considerable attention, owing to its g-wave spin splitting and high Néel temperature. In this article, we employed mechanical point-contact spectroscopy (MPCS) with superconducting Nb tips to probe the Andreev reflection on CrSb single crystals along three principal crystallographic orientations. The extracted momentum-dependent spin polarizations are approximately 73.4% for the (0001) plane, 67.9% for the (-1-120) plane, and 61.9% for the (10-10) plane, respectively, distinct from conventional antiferromagnets. Furthermore, conductance spectra from spatial line-scans on the sample surface support the existence of altermagnetic domains with a characteristic size of 250-500 nm separated by domain-walls with width about 250 nm. These results strongly support the momentum-dependent spin polarization in altermagnetic CrSb and establish Andreev reflection as a new paradigm to probe k-dependent spin textures.

cond-mat.supr-con↗

arXiv2026

Hiding the Trees in the Forest: Building Network Covert Channels with Hash-Based Covert Carrier Filtering

As an effective anti-censorship mechanism, network covert channels can provide data privacy protection and ensure communication security. However, the covertness of existing network covert channels primarily depends on the secrecy of their covert algorithms. With the increasing depth of research in this field, the difficulty of breaking such algorithms has gradually decreased. Once the algorithm is exposed, the network covert channel can be easily detected by adversaries. To address this issue, this paper proposes a covert carrier filtering strategy based on the hash. In this strategy, a key-dependent filtering rule is introduced during the construction of the network covert channel, enabling the communicating parties to randomly and dynamically filter a sparse subset from the carrier set as the covert carrier set. This strategy not only enhances the randomness of carrier selection but also tightly couples the covertness of the network covert channel with the security of the key. We employ machine learning-based traffic analysis methods to experimentally validate the strategy in two types of network covert channels: network storage and timing covert channels. The experimental results demonstrate that the proposed strategy significantly improves the detection resistance of network covert channels. When the filter key size exceeds six bits, the impact on the detection effect of the classifier becomes quite significant. Furthermore, the processing delay for a single packet is less than 8 $μs$, indicating the feasibility of deploying the proposed strategy in high-speed network environments.

cs.CR↗

arXiv2026

Optimal Illumination via Joint Movement and Phase Optimization for Movable Antenna-RIS Configuration

Reconfigurable intelligent surfaces (RIS) enable programmable control of wireless propagation but remain vulnerable to persistent deep fades in static deployments. This paper introduces a Movable Antenna-enhanced RIS (MA-RIS) architecture where antenna elements physically reposition to sample independent spatial channels, enabling mobility-induced diversity. We model antenna motion using a Stochastic Differential Equation (SDE) framework capturing controlled drift and environmental diffusion. It^o calculus-based analysis characterizes steady-state antenna distributions, spatial decorrelation, and outage probability, revealing fundamental trade-offs between control strength and mobility randomness. To maximize long-term SNR while accounting for control overhead, we propose an overhead-aware Two-timescale framework separating slow antenna trajectory control from fast phase adaptation. The stochastic optimal control problem is solved via predictive approximation of the Hamilton-Jacobi-Bellman (HJB) formulation, enabling real-time implementation. Simulations validate theoretical predictions: the Two-timescale strategy achieves up to 36 dB steady-state SNR with remarkable stability, outperforming position-only control by up to 15 dB and uncontrolled baselines by over 30 dB. Despite experiencing a lower SNR than Active RIS, the proposed approach delivers up to 16 times higher energy efficiency (EE) across varying system scales, establishing a new paradigm of mobility-enabled channel adaptation for resilient wireless systems.

eess.SP↗

arXiv2026

An Efficient, Reliable and Observable Collective Communication Library in Large-scale GPU Training Clusters

Large-scale LLM training requires collective communication libraries to exchange data among distributed GPUs. As a company dedicated to building and operating large-scale GPU training clusters, we encounter several practical limitations of NCCL in production, including 1) SM competition between computation and communication, 2) expensive restart costs under link failures, and 3) insufficient observability of transient collective communication anomalies. To address these challenges, we propose VCCL, an efficient, reliable, and observable collective communication library in large-scale GPU training clusters. VCCL removes SM-consuming P2P kernels by moving intra-node data movement and stream dependency enforcement to CPU threads and GPU copy engines. VCCL also introduces a primary-backup QP mechanism to tolerate frequent NIC port failures, and designs a window-based monitor to observe network anomalies at O(μs) level. We opensource VCCL and deploy it in production training clusters for several months. Compared with NCCL, VCCL improves training throughput by up to 5.28% and reduces massive GPU resource wastage through runtime fault tolerance and finegrained monitor. We also share experience and lessons we learned during the deployment of VCCL in large-scale clusters.

cs.DC↗

arXiv2026

Coexistence of topologically nontrivial and trivial insulating states in topological Anderson Chern insulator

The interplay between disorder and topology has become a central theme in condensed matter physics. Disorder can not only destroy topological phases but also induce them, as exemplified by the topological Anderson insulator (TAI). Here we show that, in close analogy, disorder can drive the clean-limit, time-reversal-broken(T-broken) quantum spin Hall state of ferromagnetic(FM) monolayer MnBi4Te7 into a quantum anomalous Hall phase, which was called topological Anderson Chern insulator (TACI). Using density functional theory (DFT) and nonequilibrium Green's func tion (NEGF) calculations in the presence of disorder, we identify disorder induced phases-including T-broken TAI, TACI, Normal insulator, etc., then construct a comprehensive phase diagram. To discriminate multiple phases in the strong disorder regime, we further use the density of states computed within the self-consistent Born approximation (SCBA), which in particular distinguishes gapped and ungapped topological phases. We find that the two effective band inversions of Hamiltonian are suppressed at distinct critical disorder strengths; the survival of a single inversion over a finite disorder window stabilizes the TACI. Remarkably, at strong disorder, we further propose a zero Hall plateau insulating state characterized by an insulating bulk and edge channels subject to diffusive scattering that can coexist with the TACI. This behavior is distinct from a conventional band-gap Chern insulator and provides a clear experimental signature.

cond-mat.dis-nn↗

arXiv2026

HetCCL: Enabling Collective Communication For Mixed-Vendor Heterogeneous Clusters

Training Large Language Models (LLMs) on heterogeneous clusters presents significant challenges for collective communication, as hardware from multiple vendors introduces diverse network and computational characteristics. Existing collective communication frameworks (e.g., NCCL, RCCL) designed for homogeneous environments fail to address mixed-hardware setups, while communication libraries with heterogeneous support (e.g., Gloo, OpenMPI) incur heavy overhead in the data path. This paper presents HetCCL, a framework that enables heterogeneous collective communication by efficient P2P transport across heterogeneous devices (e.g., GPUs), eliminating the host-device memory copy overhead while offloading the control to the CPUs. For combining collectives (e.g., AllReduce, ReduceScatter), HetCCL introduces a border-communicator mechanism that achieves vendor independence by using the intrinsic reduction in the combining collectives in vendor collective communication libraries. With efficient heterogeneous P2P transport and portable reduction mechanism, HetCCL proposes a hierarchical topology abstraction for heterogeneous clusters, dissecting collective communication into cluster-level primitives that guarantee optimal cross-cluster data transfer volume and optimal bandwidth utilization. We implement HetCCL with 4 different vendor support and evaluate it in 4 heterogeneous settings with benchmarks and end-to-end LLM tasks. Our evaluation shows that HetCCL achieves 17-19x higher bandwidth than Gloo in heterogeneous communications, and speeds up end-to-end training by up to 16.9% in the per-step-time.

cs.NI↗

arXiv2026

SCOPE: Simulating Cross-game Operations in Playable Environments for FPS World Models

Interactive world models for first-person shooter (FPS) games must resolve high-frequency overlapping control signals at every frame without disrupting unaffected regions. Existing methods inject actions globally and train on single titles, failing under dense FPS inputs. We observe that FPS actions are spatially selective: discrete events such as firing or reloading affect only a localized region around the weapon (the scope), while continuous camera and movement signals govern stable surroundings. We propose SCOPE, which inserts a conditioning module into each transformer block of a pretrained video diffusion model. It reshapes features into per-pixel temporal sequences so that each position computes its action response from local visual content. This separates in-scope effects from out-of-scope generation without segmentation labels. We also introduce CrossFPS, the first multi-game FPS dataset with frame-aligned action telemetry. It comprises 69K clips from 7 titles with 10-DoF controller signals, curated to remove gameplay bias. The model learns general visual-to-action mappings rather than game-specific patterns, enabling zero-shot transfer to unseen scenes. Experiments confirm strong action responsiveness, precise scope separation, and effective cross-game generalization.

cs.CV↗

arXiv2026

DynSess: Dynamic Session-Level Evaluation and Optimization Framework for Role-Playing Agents

Role-playing with large language models is fundamentally a session-level task, requiring agents to sustain character identity and interaction quality across extended multi-turn conversations. Yet existing evaluation and optimization methods remain largely turn-level, failing to capture long-horizon quality. We propose DynSess, a unified session-level framework for role-playing agents. DynSess-Eval scores complete dialogue sessions via rubrics targeting long-horizon behaviors. Leveraging its session-level rewards, we construct high-quality training trajectories through multi-turn lookahead search and train DynSess-Character with two complementary variants: DSPO (off-policy) and GSRPO (on-policy). Experiments show that DynSess-Eval aligns with human judgments substantially better than prior evaluators, and blind human evaluation further shows that DynSess-Character matches the strongest character model despite using substantially fewer parameters, while maintaining strong role consistency and interactive ability. Our dataset and code will be released to facilitate future research.

cs.CL↗

arXiv2026

SketchAssist: A Practical Assistant for Semantic Edits and Precise Local Redrawing

Sketch editing requires jointly handling high-level semantic changes and precise local redrawing, a combination that is particularly challenging for sparse, style-sensitive line art. Unlike natural images, sketches rely on minimal visual cues, making it difficult for existing methods to reconcile global semantic modifications with fine-grained structural control while preserving overall coherence. We present SketchAssist, an interactive sketch assistant that unifies instruction-guided editing with line-guided region redrawing, enabling efficient and controllable sketch manipulation while preserving overall composition. To support this task, we introduce a controllable data generation pipeline that constructs structured edit sequences with precise attribute variations and maintains structural alignment across multi-step modifications, while expanding stylistic diversity via style-preserving transformations. Building on this data, SketchAssist adopts a unified framework based on DiT, using a multi-channel input representation to encode sketches, masks, and guidance signals within a single interface. To further handle different editing modes, we integrate a Task-guided Mixture-of-Experts (T-MoE) into LoRA layers, enabling adaptive control over semantic and structural guidance. Extensive experiments demonstrate state-of-the-art performance on both tasks, achieving strong instruction adherence and improved structural and style consistency compared to recent methods. Together, our method provide a practical and controllable solution for sketch editing.

cs.CV↗

arXiv2026

Retrieved In-Context Principles from Previous Mistakes

In-context learning (ICL) has been instrumental in adapting Large Language Models (LLMs) to downstream tasks using correct input-output examples. Recent advances have attempted to improve model performance through principles derived from mistakes, yet these approaches suffer from lack of customization and inadequate error coverage. To address these limitations, we propose Retrieved In-Context Principles (RICP), a novel teacher-student framework. In RICP, the teacher model analyzes mistakes from the student model to generate reasons and insights for preventing similar mistakes. These mistakes are clustered based on their underlying reasons for developing task-level principles, enhancing the error coverage of principles. During inference, the most relevant mistakes for each question are retrieved to create question-level principles, improving the customization of the provided guidance. RICP is orthogonal to existing prompting methods and does not require intervention from the teacher model during inference. Experimental results across seven reasoning benchmarks reveal that RICP effectively enhances performance when applied to various prompting strategies.

cs.CL↗