arXiv subjects

Shuai Jiang

Publications and source records attributed to Shuai Jiang.

At least 19 recordsLinked to original sources

arXiv2026

Sub-Pixel Affine Registration of Space Debris Images via the Radon Point Spread Function

Inter-frame affine misalignment caused by platform jitter and attitude adjustments poses a fundamental challenge for multi-frame analysis of point targets in optical surveillance. Conventional registration methods rely on spatial intensity correlations or distinctive image features, both of which are largely absent in low-signal-to-noise-ratio point target imagery. We introduce the Radon Point Spread Function (RPSF) to characterize point targets in the Radon-transformed domain, and derive a closed-form framework that jointly estimates inter-frame translation and rotation from as few as four scalar RPSF samples per frame pair. The method requires no iterative optimization, feature extraction or interpolation, which is suitable for resource-constrained onboard processing. Simulation results confirm sub-pixel translation accuracy and a mean rotation error of 0.2556{\deg} at 1{\deg} Radon angular resolution. Validation on five real space debris datasets including both ground-based and in-orbit observations yields a mean calibration error below 0.5 pixels, substantially exceeding the precision required for reliable multi-frame processing.

cs.CV↗

arXiv2026

SafeBuild-Bench: A Temporal-Robust Construction Safety Benchmark with Graph-Enhanced Data Mining

Construction-safety models must handle concrete deployment risks, such as a worker standing near a scaffold edge without guardrails, rather than only recognize common objects in curated images. Yet real inspection archives are redundant, long-tailed, and collected across changing sites and months. We introduce SafeBuild-Bench, a metadata-driven benchmark for evaluating multimodal large language models on construction safety under realistic temporal and site variation. It is mined from 100K+ industrial image-text records and contains 3,314 task instances from over 3,000 expert-verified images, covering multiple-choice hazard identification and free-form hazard description. To make expert verification scalable, we develop GEMS, a graph-enhanced multimodal selection pipeline that combines a proxy-model confusion signal with graph-based diversity to identify informative candidates from redundant streams. On public instruction-tuning data, GEMS-selected subsets preserve robustness-oriented performance under small data budgets. On SafeBuild-Bench, current MLLMs remain far from reliable construction-safety understanding, with the best overall score near 60. We release the benchmark, evaluation scripts, and GEMS codebase at https://github.com/safebuild/gems.

cs.CV↗

arXiv2026

Emerging Amines reshape the paradigm of urban atmospheric particle formation

New particle formation (NPF) contributes to more than half of global aerosol number concentrations, with profound implications for human health and climate change. Observational studies have shown that the frequency of NPF events in urban Beijing during summer exceeds the global average. The prevailing paradigm attributes urban NPF primarily to sulfuric acid-base nucleation involving dimethylamine (DMA). However, recent field measurements in summer urban Beijing have identified several emerging amines emitted from carbon capture processes, including monoethanolamine (MEA), piperazine (PZ), diethanolamine (DEA) and N-methyldiethanolamine (MDEA), in addition to DMA. Here, we systematically evaluate the contributions of sulfuric acid-amine nucleation pathways to urban NPF. We found that emerging amines particularly DEA and PZ, can dominate nucleation pathways under polluted urban conditions, surpassing the contribution of DMA. These findings suggest that the current universal paradigm of urban nucleation should be revisited to explicitly account for the role of emerging amines. Moreover, emerging amine-mediated NPF will become increasingly important in the context of future co-control policies for air pollution and carbon reduction.

physics.atm-clus↗

arXiv2026

Muon in Vision Transformers: Optimizer-Recipe Interactions and Gradient Spectra

Muon is a recently developed matrix-aware optimizer that has shown strong results in transformer training, but its behavior in vision transformers (ViTs) is not yet well understood. We study Muon for ViT training, largely on ImageNet-100 and Pl@ntNet-300K, comparing against AdamW under standard vision recipes involving mixup, cutmix, smoothing, and random augmentation and erasing. Muon consistently outperforms AdamW, with especially large gains on long-tailed Pl@ntNet macro top-1. These gains are also recipe-dependent, where Muon benefits much more than AdamW from advanced and significant data augmentation techniques. To understand this interaction, we analyze the singular-value structure of matrix gradients throughout the ViT. Within Muon training runs, removing heavy data augmentation induces a late-training spectral concentration and mode collapse in gradient matrices, primarily in deep MLP-down blocks. Under a fixed "full" augmentation recipe, the clearest Muon-AdamW contrast appears instead in QKV gradients, where AdamW gradient energy remains concentrated in a much narrower basis while Muon spreads energy across substantially more singular modes. Muon in ViTs is therefore best understood as an optimizer-recipe interaction. Under a fixed recipe, Muon differs from AdamW most clearly in attention projections, where its gradients consist of a broader spectral basis. Within Muon, a full training recipe is important for preventing late spectral concentration and mode collapse in deep feedforward blocks. We further demonstrate efficacy in training ViTs on image segmentation and masked autoencoder models, where Muon outperforms AdamW in all settings considered.

cs.LG↗

arXiv2026

Temporal Preference Concepts and their Functions in a Large Language Model

Large Language Models (LLMs) are increasingly being deployed to make decisions that require trading off near-term gains against long-term consequences, yet little is known about how they internally represent or resolve these tradeoffs. In this work, we causally localize an underlying subgraph for temporal preference in a distilled LLM (Qwen3-4B-Instruct-2507), identifying mid-to-upper-layer nodes through converging evidence from gradient-based attribution and activation patching. We find that the geometry of time horizon is encoded in the residual stream at the expected localized layers. A behavioral analysis reveals that unintervened LLMs discount the future several times less steeply than humans, yet this preference is unstable across contexts, motivating explicit control rather than implicit reliance on training. Finally, we find suggestive evidence that steering vectors can shift temporal preference. Our work demonstrates how mechanistic interpretability can bring us closer to reliable control over how LLMs plan and reason

cs.LG↗

arXiv2026

Aligning What EEG Can See: Structural Representations for Brain-Vision Matching

Visual decoding from electroencephalography (EEG) has emerged as a highly promising avenue for non-invasive brain-computer interfaces (BCIs). Existing EEG-based decoding methods predominantly align brain signals with the final-layer semantic embeddings of deep visual models. However, relying on these highly abstracted embeddings inevitably leads to severe cross-modal information mismatch. In this work, we introduce the concept of Neural Visibility and accordingly propose the EEG-Visible Layer Selection Strategy, aligning EEG signals with intermediate visual layers to minimize this mismatch. Furthermore, to accommodate the multi-stage nature of human visual processing, we propose a novel Hierarchically Complementary Fusion (HCF) framework that jointly integrates visual representations from different hierarchical levels. Extensive experiments demonstrate that our method achieves state-of-the-art performance, reaching an 84.6% accuracy (+21.4%) on zero-shot visual decoding on the THINGS-EEG dataset. Moreover, our method achieves up to a 129.8% performance gain across diverse EEG baselines, demonstrating its robust generalizability.

cs.CV↗

arXiv2026

OpenVTON-Bench: A Large-Scale High-Resolution Benchmark for Controllable Virtual Try-On Evaluation

Recent advances in diffusion models have significantly elevated the visual fidelity of Virtual Try-On (VTON) systems, yet reliable evaluation remains a persistent bottleneck. Traditional metrics struggle to quantify fine-grained texture details and semantic consistency, while existing datasets fail to meet commercial standards in scale and diversity. We present OpenVTON-Bench, a large-scale benchmark comprising approximately 100K high-resolution image pairs (up to $1536 \times 1536$). The dataset is constructed using DINOv3-based hierarchical clustering for semantically balanced sampling and Gemini-powered dense captioning, ensuring a uniform distribution across 20 fine-grained garment categories. To support reliable evaluation, we propose a multi-modal protocol that measures VTON quality along five interpretable dimensions: background consistency, identity fidelity, texture fidelity, shape plausibility, and overall realism. The protocol integrates VLM-based semantic reasoning with a novel Multi-Scale Representation Metric based on SAM3 segmentation and morphological erosion, enabling the separation of boundary alignment errors from internal texture artifacts. Experimental results show strong agreement with human judgments (Kendall's $\tau$ of 0.833 vs. 0.611 for SSIM), establishing a robust benchmark for VTON evaluation.

cs.CV↗

arXiv2026

Layer-Parallel Training for Transformers

We present a new training methodology for transformers using a multilevel, layer-parallel approach. Through a neural ODE formulation of transformers, our application of a multilevel parallel-in-time algorithm for the forward and backpropagation phases of training achieves parallel acceleration over the layer dimension. This dramatically enhances parallel scalability as the network depth increases, which is particularly useful for increasingly large foundational models. However, achieving this introduces errors that cause systematic bias in the gradients, which in turn reduces convergence when closer to the minima. We develop an algorithm to detect this critical transition and either switch to serial training or systematically increase the accuracy of layer-parallel training. Results, including BERT, GPT2, ViT, and machine translation architectures, demonstrate parallel-acceleration as well as accuracy commensurate with serial pre-training while fine-tuning is unaffected.

cs.LG↗

arXiv2026

On the Convergence Behavior of Preconditioned Gradient Descent Toward the Rich Learning Regime

Spectral bias, the tendency of neural networks to learn low frequencies first, can be both a blessing and a curse. While it enhances the generalization capabilities by suppressing high-frequency noise, it can be a limitation in scientific tasks that require capturing fine-scale structures. The delayed generalization phenomenon known as grokking is another barrier to rapid training of neural networks. Grokking has been hypothesized to arise as learning transitions from the NTK to the feature-rich regime. This paper explores the impact of preconditioned gradient descent (PGD), such as Gauss-Newton, on spectral bias and grokking phenomena. We demonstrate through theoretical and empirical results how PGD can mitigate issues associated with spectral bias. Additionally, building on the rich learning regime grokking hypothesis, we study how PGD can be used to reduce delays associated with grokking. Our conjecture is that PGD, without the impediment of spectral bias, enables uniform exploration of the parameter space in the NTK regime. Our experimental results confirm this prediction, providing strong evidence that grokking represents a transitional behavior between the lazy regime characterized by the NTK and the rich regime. These findings deepen our understanding of the interplay between optimization dynamics, spectral bias, and the phases of neural network learning.

cs.LG↗

arXiv2025

Automated Neural Architecture Design for Industrial Defect Detection

Industrial surface defect detection (SDD) is critical for ensuring product quality and manufacturing reliability. Due to the diverse shapes and sizes of surface defects, SDD faces two main challenges: intraclass difference and interclass similarity. Existing methods primarily utilize manually designed models, which require extensive trial and error and often struggle to address both challenges effectively. To overcome this, we propose AutoNAD, an automated neural architecture design framework for SDD that jointly searches over convolutions, transformers, and multi-layer perceptrons. This hybrid design enables the model to capture both fine-grained local variations and long-range semantic context, addressing the two key challenges while reducing the cost of manual network design. To support efficient training of such a diverse search space, AutoNAD introduces a cross weight sharing strategy, which accelerates supernet convergence and improves subnet performance. Additionally, a searchable multi-level feature aggregation module (MFAM) is integrated to enhance multi-scale feature learning. Beyond detection accuracy, runtime efficiency is essential for industrial deployment. To this end, AutoNAD incorporates a latency-aware prior to guide the selection of efficient architectures. The effectiveness of AutoNAD is validated on three industrial defect datasets and further applied within a defect imaging and detection platform. Code is available at https://github.com/Yuxi104/AutoNAD.

cs.CV↗

arXiv2025

Hunyuan3D Studio: End-to-End AI Pipeline for Game-Ready 3D Asset Generation

The creation of high-quality 3D assets, a cornerstone of modern game development, has long been characterized by labor-intensive and specialized workflows. This paper presents Hunyuan3D Studio, an end-to-end AI-powered content creation platform designed to revolutionize the game production pipeline by automating and streamlining the generation of game-ready 3D assets. At its core, Hunyuan3D Studio integrates a suite of advanced neural modules (such as Part-level 3D Generation, Polygon Generation, Semantic UV, etc.) into a cohesive and user-friendly system. This unified framework allows for the rapid transformation of a single concept image or textual description into a fully-realized, production-quality 3D model complete with optimized geometry and high-fidelity PBR textures. We demonstrate that assets generated by Hunyuan3D Studio are not only visually compelling but also adhere to the stringent technical requirements of contemporary game engines, significantly reducing iteration time and lowering the barrier to entry for 3D content creation. By providing a seamless bridge from creative intent to technical asset, Hunyuan3D Studio represents a significant leap forward for AI-assisted workflows in game development and interactive media.

cs.CV↗

arXiv2025

Resilient Multimodal Industrial Surface Defect Detection with Uncertain Sensors Availability

Multimodal industrial surface defect detection (MISDD) aims to identify and locate defect in industrial products by fusing RGB and 3D modalities. This article focuses on modality-missing problems caused by uncertain sensors availability in MISDD. In this context, the fusion of multiple modalities encounters several troubles, including learning mode transformation and information vacancy. To this end, we first propose cross-modal prompt learning, which includes: i) the cross-modal consistency prompt serves the establishment of information consistency of dual visual modalities; ii) the modality-specific prompt is inserted to adapt different input patterns; iii) the missing-aware prompt is attached to compensate for the information vacancy caused by dynamic modalities-missing. In addition, we propose symmetric contrastive learning, which utilizes text modality as a bridge for fusion of dual vision modalities. Specifically, a paired antithetical text prompt is designed to generate binary text semantics, and triple-modal contrastive pre-training is offered to accomplish multimodal learning. Experiment results show that our proposed method achieves 73.83% I-AUROC and 93.05% P-AUROC with a total missing rate 0.7 for RGB and 3D modalities (exceeding state-of-the-art methods 3.84% and 5.58% respectively), and outperforms existing approaches to varying degrees under different missing types and rates. The source code will be available at https://github.com/SvyJ/MISDD-MM.

cs.CV↗

arXiv2025

Interpretability and Generalization Bounds for Learning Spatial Physics

While there are many applications of ML to scientific problems that look promising, visuals can be deceiving. Using numerical analysis techniques, we rigorously quantify the accuracy, convergence rates, and generalization bounds of certain ML models applied to linear differential equations for parameter discovery or solution finding. Beyond the quantity and discretization of data, we identify that the function space of the data is critical to the generalization of the model. A similar lack of generalization is empirically demonstrated for commonly used models, including physics-specific techniques. Counterintuitively, we find that different classes of models can exhibit opposing generalization behaviors. Based on our theoretical analysis, we also introduce a new mechanistic interpretability lens on scientific models whereby Green's function representations can be extracted from the weights of black-box models. Our results inform a new cross-validation technique for measuring generalization in physical systems, which can serve as a benchmark.

cs.LG↗

arXiv2025

Transformer-Based Representation Learning for Robust Gene Expression Modeling and Cancer Prognosis

Transformer-based models have achieved remarkable success in natural language and vision tasks, but their application to gene expression analysis remains limited due to data sparsity, high dimensionality, and missing values. We present GexBERT, a transformer-based autoencoder framework for robust representation learning of gene expression data. GexBERT learns context-aware gene embeddings by pretraining on large-scale transcriptomic profiles with a masking and restoration objective that captures co-expression relationships among thousands of genes. We evaluate GexBERT across three critical tasks in cancer research: pan-cancer classification, cancer-specific survival prediction, and missing value imputation. GexBERT achieves state-of-the-art classification accuracy from limited gene subsets, improves survival prediction by restoring expression of prognostic anchor genes, and outperforms conventional imputation methods under high missingness. Furthermore, its attention-based interpretability reveals biologically meaningful gene patterns across cancer types. These findings demonstrate the utility of GexBERT as a scalable and effective tool for gene expression modeling, with translational potential in settings where gene coverage is limited or incomplete.

cs.LG↗

arXiv2024

Improving Colorectal Cancer Screening and Risk Assessment through Predictive Modeling on Medical Images and Records

Colonoscopy screening effectively identifies and removes polyps before they progress to colorectal cancer (CRC), but current follow-up guidelines rely primarily on histopathological features, overlooking other important CRC risk factors. Variability in polyp characterization among pathologists also hinders consistent surveillance decisions. Advances in digital pathology and deep learning enable the integration of pathology slides and medical records for more accurate CRC risk prediction. Using data from the New Hampshire Colonoscopy Registry, including longitudinal follow-up, we adapted a transformer-based model for histopathology image analysis to predict 5-year CRC risk. We further explored multi-modal fusion strategies to combine clinical records with deep learning-derived image features. Training the model to predict intermediate clinical variables improved 5-year CRC risk prediction (AUC = 0.630) compared to direct prediction (AUC = 0.615, p = 0.013). Incorporating both imaging and non-imaging data, without requiring manual slide review, further improved performance (AUC = 0.674) compared to traditional features from colonoscopy and microscopy reports (AUC = 0.655, p = 0.001). These results highlight the value of integrating diverse data modalities with computational methods to enhance CRC risk stratification.

cs.CV↗

arXiv2024

Generative AI on SpectrumNet: An Open Benchmark of Multiband 3D Radio Maps

Radio map is an efficient demonstration for visually displaying the wireless signal coverage within a certain region. It has been considered to be increasingly helpful for the future sixth generation (6G) of wireless networks, as wireless nodes are becoming more crowded and complicated. However, the construction of high resolution radio map is very challenging due to the sparse sampling in practical systems. Generative artificial intelligence (AI), which is capable to create synthetic data to fill in gaps in real-world measurements, is an effective technique to construct high precision radio maps. Currently, generative models for radio map construction are trained with two-dimension (2D) single band radio maps in urban scenario, which has poor generalization in diverse terrain scenarios, spectrum bands, and heights. To tackle this problem, we provide a multiband three-dimension (3D) radio map dataset with consideration of terrain and climate information, named SpectrumNet. It is the largest radio map dataset in terms of dimensions and scale, which contains the radio map of 3 spacial dimensions, 5 frequency bands, 11 terrain scenarios, and 3 climate scenarios. We introduce the parameters and settings for the SpectrumNet dataset generation, and evaluate three baseline methods for radio map construction based on the SpectrumNet dataset. Experiments show the necessity of the SpectrumNet dataset for training models with strong generalization in spacial, frequency, and scenario domains. Future works on the SpectrumNet dataset are also discussed, including the dataset expansion and calibration, as well as the extended studies on generative models for radio map construction based on the SpectrumNet dataset.

eess.SP↗

arXiv2024

A Structure-Preserving Domain Decomposition Method for Data-Driven Modeling

We present a domain decomposition strategy for developing structure-preserving finite element discretizations from data when exact governing equations are unknown. On subdomains, trainable Whitney form elements are used to identify structure-preserving models from data, providing a Dirichlet-to-Neumann map which may be used to globally construct a mortar method. The reduced-order local elements may be trained offline to reproduce high-fidelity Dirichlet data in cases where first principles model derivation is either intractable, unknown, or computationally prohibitive. In such cases, particular care must be taken to preserve structure on both local and mortar levels without knowledge of the governing equations, as well as to ensure well-posedness and stability of the resulting monolithic data-driven system. This strategy provides a flexible means of both scaling to large systems and treating complex geometries, and is particularly attractive for multiscale problems with complex microstructure geometry. While consistency is traditionally obtained in finite element methods via quasi-optimality results and the Bramble-Hilbert lemma as the local element diameter $h\rightarrow0$, our analysis establishes notions of accuracy and stability for finite h with accuracy coming from matching data. Numerical experiments and analysis establish properties for $H(\operatorname{div})$ problems in small data limits ($\mathcal{O}(1)$ reference solutions).

math.NA↗

arXiv2024

MindShot: A Few-Shot Brain Decoding Framework via Transferring Cross-Subject Prior and Distilling Frequency Domain Knowledge

Aiming to reconstruct visual stimuli from brain signals, brain decoding has recently made significant progress using functional magnetic resonance imaging (fMRI). However, it still has challenging issues such as substantial individual differences and high data collection costs. To simplify these problems, most methods adopt the per-subject-per-model paradigm, but this greatly limits their applications. In this paper, we design a few-shot brain decoding setting specifically for potential clinical scenarios and propose a novel two-stage decoding framework named MindShot, comprising a Multi-Subject Pretraining (MSP) stage and Fourier-based cross-subject Knowledge Distillation (FKD) stage. Firstly, a MSP framework based on multi-modal contrastive learning is constructed to mine the cross-subject prior. Secondly, the FKD is presented to decrease inter-individual differences while improving the decoding adaptability to new individuals. Our approach achieves high semantic fidelity in visual reconstruction on the largest dataset and has the potential to reduce scanning time by up to 99%. Remarkably, MindShot achieves a CLIP accuracy of 83.6% using only 1.8% of the fMRI-image pairs, surpassing the 77.4% accuracy of the method trained on the entire NSD dataset. This makes it feasible to train large-scale brain decoding frameworks that require less data, facilitating practical applications. The code is available at https://github.com/JSinBUPT/MindShot.

cs.CV↗