arXiv subjects

Weijian Liu

Publications and source records attributed to Weijian Liu.

17 recordsLinked to original sources

arXiv2026

CrossAtlas: Evaluating Projection Techniques for Spatial Referencing in Cross-Reality Collaboration

Cross-reality collaboration increasingly connects immersive and desktop users within synchronized workspaces, yet little is known about how bidirectional projection techniques between immersive 3D layouts and desktop 2D views influence communication. Spatial referencing depends on shared spatial understanding, but different mappings preserve and distort geometric relationships in different ways, altering perceived adjacency, orientation, and coverage across collaborators' views. We present CrossAtlas, a synchronized PC-VR collaboration platform that integrates multiple bidirectional projection techniques, including three planar projection variants and equirectangular, a spherical projection variant, across layouts of varying curvature. In a controlled study with 24 dyads, collaborators completed spatial referencing tasks under different projection-layout conditions while we collected performance and subjective measures. Our results show that projection choice strongly shaped collaboration, with the spherical variant often outperforming planar projections and remaining robust across object layouts.

cs.HC↗

arXiv2026

Direct Model State Migration for Elastic Training of Large Language Models

Large language model (LLM) training shall adapt to dynamic resources in shared clusters to tackle the elasticity, including passive preemption and optimistic scaling. State migration across device sets is required when altering the hybrid-parallel configuration due to dynamic resources. Existing solutions rely on checkpoint-based mechanisms, which persist complete states to storage for resuming with re-assigned resources, forcing all GPUs to stall when transferring model states. Despite optimization efforts, checkpoint-based solutions incur prohibitive latency due to data movement across memory hierarchies. We propose ETC, a checkpoint-free state migration framework for elastic hybrid-parallel LLM training. We exploits the state locality to minimize inter-GPU data movement, replacing storage persistence with direct peer-to-peer communication. Besides, we eliminate node fragmentation through communication coalescing. Integrated with Megatron-LM, ETC reduces migration overhead by 2.33$\times$ to 6.37$\times$ compared to checkpoint-based solutions across diverse parallel configurations. By enabling efficient migration, ETC unlocks practical elastic training in production environments.

cs.DC↗

arXiv2026

JanusPipe: Efficient Pipeline Parallel Training for Machine Learning Interatomic Potentials

Discovering atom-level phenomena requires molecular dynamics (MD) simulations with ab initio accuracy. Machine learning interatomic potentials (MLIPs) enable stable, high-accuracy MD simulations, and their models exhibit scaling-law trends similar to large language models. However, the lack of scalable and efficient distributed training systems for conservative MLIPs makes them difficult to scale. This is because conservative MLIPs inherently follow a double-backward execution pattern, which involves computing gradients during the forward pass. This pattern creates a mismatch with existing distributed training systems, especially for pipeline parallelism. Therefore, we present JanusPipe, an efficient 3D-parallel (PP/DP/GP) training system tailored for conservative MLIPs. It integrates SymFold to enable memory-efficient pipeline parallelism for conservative MLIPs, and WaveK to reduce pipeline bubbles by balancing the four-phase compute time. Experimental results on 32 GPUs show that JanusPipe improves throughput by $1.51\times$ and $1.45\times$ on average over 1F1B and Hanayo, respectively.

cs.DC↗

arXiv2026

Adaptive Subspace Signal Detection and Performance Analysis in Nonzero-Mean Clutter

To solve the problem of detecting subspace signals in nonzero-mean clutter, we propose adaptive detectors, based on the strategies of generalized likelihood ratio test (GLRT), Rao test, Wald test, gradient test, and Durbin test. The results show that the detectors based on GLRT, Rao and Wald are structurally consistent with the subspace detectors in zero-means clutter. The analytic expressions for the probability of detection (PD) and probability of false alarm (PFA) of each detector are derived, and two major performance differences in the nonzero-mean clutter scenario are revealed. One is the loss of degree of freedom (DOF), which is reduced by 1 compared with the zero-mean clutter scenario. The second is the loss of signal-to-clutter (SCR) ratio. Simulation and measured data verify the effectiveness of the proposed detectors and demonstrate their practical value in real-world radar systems.

stat.OT↗

arXiv2026

Breaking the Training Barrier of Billion-Parameter Universal Machine Learning Interatomic Potentials

Universal Machine Learning Interatomic Potentials (uMLIPs), pre-trained on massively diverse datasets encompassing inorganic materials and organic molecules across the entire periodic table, serve as foundational models for quantum-accurate physical simulations. However, uMLIP training requires second-order derivatives, which lack corresponding parallel training frameworks; moreover, scaling to the billion-parameter regime causes explosive growth in computation and communication overhead, making its training a tremendous challenge. We introduce MatRIS-MoE, a billion-parameter Mixture-of-Experts model built upon invariant architecture, and {Janus}, a pioneering high-dimensional distributed training framework for uMLIPs with hardware-aware optimizations. Deployed across two Exascale supercomputers, our code attains a peak performance of 1.2/1.0 EFLOPS (24\%/{35.5\%} of theoretical peak) in single precision at over 90\% parallel efficiency, compressing the training of billion-parameter uMLIPs from weeks to hours. This work establishes a new high-water mark for AI-for-Science (AI4S) foundation models at Exascale and provides essential infrastructure for rapid scientific discovery.

cs.DC↗

arXiv2025

PR-CapsNet: Pseudo-Riemannian Capsule Network with Adaptive Curvature Routing for Graph Learning

Capsule Networks (CapsNets) show exceptional graph representation capacity via dynamic routing and vectorized hierarchical representations, but they model the complex geometries of real\-world graphs poorly by fixed\-curvature space due to the inherent geodesical disconnectedness issues, leading to suboptimal performance. Recent works find that non\-Euclidean pseudo\-Riemannian manifolds provide specific inductive biases for embedding graph data, but how to leverage them to improve CapsNets is still underexplored. Here, we extend the Euclidean capsule routing into geodesically disconnected pseudo\-Riemannian manifolds and derive a Pseudo\-Riemannian Capsule Network (PR\-CapsNet), which models data in pseudo\-Riemannian manifolds of adaptive curvature, for graph representation learning. Specifically, PR\-CapsNet enhances the CapsNet with Adaptive Pseudo\-Riemannian Tangent Space Routing by utilizing pseudo\-Riemannian geometry. Unlike single\-curvature or subspace\-partitioning methods, PR\-CapsNet concurrently models hierarchical and cluster or cyclic graph structures via its versatile pseudo\-Riemannian metric. It first deploys Pseudo\-Riemannian Tangent Space Routing to decompose capsule states into spherical\-temporal and Euclidean\-spatial subspaces with diffeomorphic transformations. Then, an Adaptive Curvature Routing is developed to adaptively fuse features from different curvature spaces for complex graphs via a learnable curvature tensor with geometric attention from local manifold properties. Finally, a geometric properties\-preserved Pseudo\-Riemannian Capsule Classifier is developed to project capsule embeddings to tangent spaces and use curvature\-weighted softmax for classification. Extensive experiments on node and graph classification benchmarks show PR\-CapsNet outperforms SOTA models, validating PR\-CapsNet's strong representation power for complex graph structures.

cs.LG↗

arXiv2025

LitForager: Exploring Multimodal Literature Foraging Strategies in Immersive Sensemaking

Exploring and comprehending relevant academic literature is a vital yet challenging task for researchers, especially given the rapid expansion in research publications. This task fundamentally involves sensemaking - interpreting complex, scattered information sources to build understanding. While emerging immersive analytics tools have shown cognitive benefits like enhanced spatial memory and reduced mental load, they predominantly focus on information synthesis (e.g., organizing known documents). In contrast, the equally important information foraging phase - discovering and gathering relevant literature - remains underexplored within immersive environments, hindering a complete sensemaking workflow. To bridge this gap, we introduce LitForager, an interactive literature exploration tool designed to facilitate information foraging of research literature within an immersive sensemaking workflow using network-based visualizations and multimodal interactions. Developed with WebXR and informed by a formative study with researchers, LitForager supports exploration guidance, spatial organization, and seamless transition through a 3D literature network. An observational user study with 15 researchers demonstrated LitForager's effectiveness in supporting fluid foraging strategies and spatial sensemaking through its multimodal interface.

cs.HC↗

arXiv2025

Statistical Performance of Generalized Direction Detectors with Known Spatial Steering Vector

The generalized direction detection (GDD) problem involves determining the presence of a signal of interest within matrix-valued data, where the row and column spaces of the signal (if present) are known, but the speciffc coordinates are unknown. Many detectors have been proposed for GDD, yet there is a lack of analytical results regarding their statistical detection performance. This paper presents a theoretical analysis of two adaptive detectors for GDD in scenarios with known spatial steering vectors. Speciffcally, we establish their statistical distributions and develop closed-form expressions for both detection probability (PD) and false alarm probability (PFA). Simulation experiments are carried out to validate the theoretical results, demonstrating good agreement between theoretical and simulated results.

stat.AP↗

arXiv2025

Bayesian Rao test for distributed target detection in interference and noise with limited training data

This paper has studied the problem of detecting a range-spread target in interference and noise when the number of training data is limited. The interference is located within a certain subspace with an unknown coordinate, while the noise follows a Gaussian distribution with an unknown covariance matrix. We concentrate on the scenarios where the training data are limited and employ a Bayesian framework to ffnd a solution. Speciffcally, the covariance matrix is assumed to follow an inverse Wishart distribution. Then, we introduce the Bayesian detector according to the Rao test, which, demonstrated by both simulation experiment and real data, has superior detection performance to the existing detectors in certain situations.

stat.ME↗

arXiv2024

Mask-RadarNet: Enhancing Transformer With Spatial-Temporal Semantic Context for Radar Object Detection in Autonomous Driving

As a cost-effective and robust technology, automotive radar has seen steady improvement during the last years, making it an appealing complement to commonly used sensors like camera and LiDAR in autonomous driving. Radio frequency data with rich semantic information are attracting more and more attention. Most current radar-based models take radio frequency image sequences as the input. However, these models heavily rely on convolutional neural networks and leave out the spatial-temporal semantic context during the encoding stage. To solve these problems, we propose a model called Mask-RadarNet to fully utilize the hierarchical semantic features from the input radar data. Mask-RadarNet exploits the combination of interleaved convolution and attention operations to replace the traditional architecture in transformer-based models. In addition, patch shift is introduced to the Mask-RadarNet for efficient spatial-temporal feature learning. By shifting part of patches with a specific mosaic pattern in the temporal dimension, Mask-RadarNet achieves competitive performance while reducing the computational burden of the spatial-temporal modeling. In order to capture the spatial-temporal semantic contextual information, we design the class masking attention module (CMAM) in our encoder. Moreover, a lightweight auxiliary decoder is added to our model to aggregate prior maps generated from the CMAM. Experiments on the CRUW dataset demonstrate the superiority of the proposed method to some state-of-the-art radar-based object detection algorithms. With relatively lower computational complexity and fewer parameters, the proposed Mask-RadarNet achieves higher recognition accuracy for object detection in autonomous driving.

cs.CV↗

arXiv2024

Adaptive radar detection of subspace-based distributed target in power heterogeneous clutter

This paper investigates the problem of adaptive detection of distributed targets in power heterogeneous clutter. In the considered scenario, all the data share the identical structure of clutter covariance matrix, but with varying and unknown power mismatches. To address this problem, we iteratively estimate all the unknowns, including the coordinate matrix of the target, the clutter covariance matrix, and the corresponding power mismatches, and propose three detectors based on the generalized likelihood ratio test (GLRT), Rao and the Wald tests. The results from simulated and real data both illustrate that the detectors based on GLRT and Rao test have higher probabilities of detection (PDs) than the existing competitors. Among them, the Rao test-based detector exhibits the best overall detection performance. We also analyze the impact of the target extended dimensions, the signal subspace dimensions, and the number of training samples on the detection performance. Furthermore, simulation experiments also demonstrate that the proposed detectors have a constant false alarm rate (CFAR) property for the structure of clutter covariance matrix.

stat.ME↗

arXiv2024

LDA-MIG Detectors for Maritime Targets in Nonhomogeneous Sea Clutter

This paper deals with the problem of detecting maritime targets embedded in nonhomogeneous sea clutter, where limited number of secondary data is available due to the heterogeneity of sea clutter. A class of linear discriminant analysis (LDA)-based matrix information geometry (MIG) detectors is proposed in the supervised scenario. As customary, Hermitian positive-definite (HPD) matrices are used to model the observational sample data, and the clutter covariance matrix of received dataset is estimated as geometric mean of the secondary HPD matrices. Given a set of training HPD matrices with class labels, that are elements of a higher-dimensional HPD matrix manifold, the LDA manifold projection learns a mapping from the higher-dimensional HPD matrix manifold to a lower-dimensional one subject to maximum discrimination. In the current study, the LDA manifold projection, with the cost function maximizing between-class distance while minimizing within-class distance, is formulated as an optimization problem in the Stiefel manifold. Four robust LDA-MIG detectors corresponding to different geometric measures are proposed. Numerical results based on both simulated radar clutter with interferences and real IPIX radar data show the advantage of the proposed LDA-MIG detectors against their counterparts without using LDA as well as the state-of-art maritime target detection methods in nonhomogeneous sea clutter.

eess.SP↗

arXiv2023

Detector Design and Performance Analysis for Target Detection in Subspace Interference

It is often difficult to obtain sufficient training data for adaptive signal detection, which is required to calculate the unknown noise covariance matrix. Additionally, interference is frequently present, which complicates the detecting issue. We provide a two-step method, termed interference cancellation before detection (ICBD), to address the issue of signal detection in the unknown Gaussian noise and subspace interference. The first involves projecting the test and training data to the interference-orthogonal subspace in order to suppress the interference. Utilizing traditional adaptive detector design ideas is the next stage. Due to the smaller dimension of the projected data, the ICBD-based detectors can function with little training data. The ICBD has two additional benefits over traditional detectors. Lower computational burden and proper operation with interference being in the training data are two additional benefits of ICBD-based detectors over conventional ones. We also give the statistical properties of the ICBD-based detectors and demonstrate their equivalence with the traditional ones in the special case of a large amount of training data containing no interference

stat.AP↗

arXiv2021

Detection of a rank-one signal with limited training data

In this paper, we reconsider the problem of detecting a matrix-valued rank-one signal in unknown Gaussian noise, which was previously addressed for the case of sufficient training data. We relax the above assumption to the case of limited training data. We re-derive the corresponding generalized likelihood ratio test (GLRT) and two-step GLRT (2S--GLRT) based on certain unitary transformation on the test data. It is shown that the re-derived detectors can work with low sample support. Moreover, in sample-abundant environments the re-derived GLRT is the same as the previously proposed GLRT and the re-derived 2S--GLRT has better detection performance than the previously proposed 2S--GLRT. Numerical examples are provided to demonstrate the effectiveness of the re-derived detectors.

eess.SP↗

arXiv2021

Multichannel adaptive signal detection: Basic theory and literature review

Multichannel adaptive signal detection jointly uses the test and training data to form an adaptive detector, and then make a decision on whether a target exists or not. Remarkably, the resulting adaptive detectors usually possess the constant false alarm rate (CFAR) properties, and hence no additional CFAR processing is needed. Filtering is not needed as a processing procedure either, since the function of filtering is embedded in the adaptive detector. Moreover, adaptive detection usually exhibits better detection performance than the filtering-then-CFAR detection technique. Multichannel adaptive signal detection has been more than 30 years since the first multichannel adaptive detector was proposed by Kelly in 1986. However, there are fewer overview articles on this topic. In this paper we give a tutorial overview of multichannel adaptive signal detection, with emphasis on Gaussian background. We present the main deign criteria for adaptive detectors, investigate the relationship between adaptive detection and filtering-then-CFAR detection, relationship between adaptive detectors and adaptive filters, summarize typical adaptive detectors, show numerical examples, give comprehensive literature review, and discuss some possible further research tracks.

stat.AP↗

arXiv2019

Multichannel signal detection in interference and noise when signal mismatch happens

In this paper, we consider the problem of detecting a multichannel signal in interference and noise when signal mismatch happens. We first propose two selective detectors, since their strong selectivity is preferred in some situations. However, these two detectors would not be suitable candidates if a robust detector is needed. To overcome this shortcoming, we then devise a tunable detector, which is parametrized by a non-negative scaling factor, referred to as the tunable parameter. By adjusting the tunable parameter, the proposed detector can smoothly change its capability in rejecting or robustly detecting a mismatch signal. Moreover, one selective detector and the tunable detector with an appropriate tunable parameter can provide nearly the same detection performance as existing detectors in the absence of signal mismatch. We obtain analytical expressions for the probabilities of detection (PDs) and probabilities of false alarm (PFAs) of the three proposed detectors, which are verified by Monte Carlo simulations.

eess.SP↗

arXiv2017

Sparsity-Based STAP Design Based on Alternating Direction Method with Gain/Phase Errors

We present a novel sparsity-based space-time adaptive processing (STAP) technique based on the alternating direction method to overcome the severe performance degradation caused by array gain/phase (GP) errors. The proposed algorithm reformulates the STAP problem as a joint optimization problem of the spatio-Doppler profile and GP errors in both single and multiple snapshots, and introduces a target detector using the reconstructed spatio-Doppler profiles. Simulations are conducted to illustrate the benefits of the proposed algorithm.

cs.DS↗