Searcharxiv⌕ Search

arXiv subjects

Jingtao Zhang

Publications and source records attributed to Jingtao Zhang.

15 recordsLinked to original sources
arXiv2026

Geometric Mean Pooling for Equal-Weight Multiplicative Coarse-Graining

As an alternative to the additive and extremal biases of average and max pooling, we introduce Geometric Mean Pooling (GMP), a signed pooling operator that combines the product of feature signs with the geometric mean of feature magnitudes. Motivated by local-to-global composition in quantum many-body physics, GMP retains both joint sign information and a characteristic multiplicative scale without introducing learnable pooling parameters. We show that non-overlapping hierarchical GMP preserves the corresponding global multiplicative statistic and evaluate it on synthetic sequence tasks, iterative coarse-graining, image classification, and molecular lipophilicity regression. On the synthetic tasks, GMP recovers product-based signals more accurately than average and max pooling and maintains predictive performance under the tested levels of multiplicative input noise. On image and molecular data, however, its effectiveness depends on the representation, target parameterization, and placement of local and global pooling. These results position GMP as a complementary, regime-dependent inductive bias for tasks in which equal-weight multiplicative composition is plausible, rather than as a universal replacement for standard pooling operators.

cs.LG↗
arXiv2026

BiHDTrans: binary hyperdimensional transformer for efficient multivariate time series classification

The proliferation of Internet-of-Things (IoT) devices has led to an unprecedented volume of multivariate time series (MTS) data, requiring efficient and accurate processing for timely decision-making in resource-constrained edge environments. Hyperdimensional (HD) computing, with its inherent efficiency and parallelizability, has shown promise in classification tasks but struggles to capture complex temporal patterns, while Transformers excel at sequence modeling but incur high computational and memory overhead. We introduce BiHDTrans, an efficient neurosymbolic binary hyperdimensional Transformer that integrates self-attention into the HD computing paradigm, unifying the representational efficiency of HD computing with the temporal modeling power of Transformers. Empirically, BiHDTrans outperforms state-of-the-art (SOTA) HD computing models by at least 14.47% and achieves 6.67% higher accuracy on average than SOTA binary Transformers. With hardware acceleration on FPGA, our pipelined implementation leverages the independent and identically distributed properties of high-dimensional representations, delivering 39.4 times lower inference latency than SOTA binary Transformers. Theoretical analysis shows that binarizing in holographic high-dimensional space incurs significantly less information distortion than directly binarizing neural networks, explaining BiHDTrans's superior accuracy. Furthermore, dimensionality experiments confirm that BiHDTrans remains competitive even with a 64% reduction in hyperspace dimensionality, surpassing SOTA binary Transformers by 1-2% in accuracy with 4.4 times less model size, as well as further reducing the latency by 49.8% compare to the full-dimensional baseline. Together, these contributions bridge the gap between the expressiveness of Transformers and the efficiency of HD computing, enabling accurate, scalable, and low-latency MTS classification.

cs.LG↗
arXiv2026

FaultLens: Learning Compact Behavioral Test Suites for Generated Operational Programs

Generated operational programs are often validated with either a few hand-written examples or exhaustive regression suites. The former can miss sparse boundary and interaction faults, while the latter can be unnecessarily expensive. We introduce FaultLens, a method for learning compact behavioral test suites while preserving an auditable connection to executed evidence. It executes a rich probe domain once, stores the fault-probe kill relation as a sparse outcome cache, and learns probe orderings only from earlier program generations. A fault-driven greedy component exploits known kill structure, while a mutation-independent diversity component covers probe families, cases, templates, and temporal bins. Their alternating hybrid remains useful when a new program contains a fault mechanism absent from ordering construction. We evaluate twenty generated operational policies across four environments, ten execution seeds, 1,200 measured run summaries, 2,160 controlled program transformations, and 4,120,200 executed program-probe pairs. Of 1,960 intended faulty transformations, 1,779 alter a contract or output somewhere in the finite audit domain; 200 additional controls preserve behavior. A 32-probe hybrid learned on generations 1-3 covers 576/582 (99.0%) dynamically killable faults in generations 4-5 using 1.2-2.0% of the exhaustive domain. With an entire fault family withheld from training, diversity raises scenario-family macro coverage from 84.6% to 94.9%. In a downstream deployment study, a conservative admission rule reduces severe tail regressions from 15/20 program-environment groups to 0/20. FaultLens provides a prioritized evidence mechanism, not a proof of correctness, and makes its budget, evidence source, generalization split, and misses explicit.

cs.SE↗
arXiv2026

Scale-Separated Conditioning for Style-Encoder-Free Diffusion Stylization

Reference-based diffusion stylization requires separating target geometry from transferable appearance. Existing tuning-based methods often rely on aligned content-style-target triplets or auxiliary visual encoders, which increases data cost and can transfer unintended scene structure from the style reference. We propose SEFS (Style-Encoder-Free Stylization), a style-encoder-free conditioning framework for diffusion transformers. SEFS forms style tokens from stochastic low-resolution crops of single training images. This crop bottleneck preserves local appearance statistics such as palette, stroke, texture, and material, while reducing access to global layout cues. Target content is encoded by edge and segmentation cues and fused with the noisy latent through parameter-efficient trainable projections. We add style-to-denoising re-normalization for token-statistic alignment and cross-block skip fusion for spatial detail. SEFS trains on unpaired single images; the frozen diffusion VAE is used only to place image conditions in the latent space. On artistic stylization benchmarks, SEFS improves content consistency and leakage diagnostics while retaining reference-style affinity, and ablations support the crop-resolution, re-normalization, and skip-fusion choices. The code of SEFS will be made publicly available.

cs.CV↗
arXiv2026

Threat-guided Policy-aware Scene Perturbation for Safe Autonomous Driving with Online Reinforcement Learning

Reinforcement learning (RL) has shown promising performance in autonomous driving, yet ensuring the safety of online RL policies remains challenging due to insufficient exposure to safety-critical driving scenes. The long-tailed nature of real-world traffic situations makes dangerous and rare interactions difficult to encounter through conventional sampling, limiting the ability of RL policies to learn robust safety behaviors. Existing methods improve training diversity by synthesizing challenging scenes or adversarial situations. However, these approaches typically optimize scene generation objectives separately from the evolving policy, without explicitly modeling how generated perturbations relate to the current policy's weaknesses and learning needs. In this paper, we propose Threat-guided Policy-aware Scene Perturbation (TPSP) for safe autonomous driving with online RL. TPSP introduces a policy-aware scene encoder to capture the interaction between policy behaviors and surrounding environments, enabling scene perturbation aligned with the current policy. Based on this representation, TPSP selectively perturbs critical objects rather than applying uniform modifications across the scene. Furthermore, we develop a threat-guided optimization strategy that evaluates perturbed scenes through threat-level differences between policy rollouts on original and perturbed scenes, guiding the generation of safety-critical scenes with higher training value. Comprehensive experiments demonstrate that TPSP improves safety learning efficiency, achieving strong safety performance on NAVSIM v2 with approximately 4 million kilometers of simulated driving data. Ablation studies verify that policy-aware targeted perturbations provide more informative safety-critical experiences than random or policy-unaware strategies, enabling safer driving under limited interaction budgets.

cs.AI↗
arXiv2026

Enhancing Video Physical Consistency via Role-aware Joint Training and Modality-decoupled Denoising

While modern video diffusion models excel in visual fidelity, maintaining long-range physical consistency remains a formidable challenge. Conventional pixel-reconstruction objectives mainly focus on appearance details and often fail to capture the underlying dynamics of a scene. To mitigate this, recent efforts have integrated auxiliary modalities (e.g., optical flow) to introduce physics priors via joint training with video appearance. However, these methods have three main limitations: (1) they do not distinguish the different motion patterns of different entity types; (2) joint modeling of visual and auxiliary modalities can cause capacity conflicts and weaken the pretrained visual prior; and (3) auxiliary modalities may accumulate errors during inference. To address these issues, we propose \textbf{VPT}, a fine-tuning framework for improving physical consistency in video diffusion models. VPT introduces a role-aware signal that groups entities into agents, controlled objects, passive objects, and background, so that different physical roles can be modeled more clearly. We further propose a modality-decoupled denoising strategy, where the visual and auxiliary channels are assigned independent noise levels. Together with a loss-weight decay strategy, this design makes auxiliary modalities serve as soft constraints rather than strong dependencies, mitigating recursive prediction errors during inference. We also introduce cross-step auto-guidance to further strengthen physical dynamics. Experiments show that VPT improves physical consistency while preserving visual quality, achieving relative gains of 39.4\% in SA and 17.9\% in PC on VideoPhy benchmark over Wan2.1-T2V-1.3B, and consistent improvements on VideoPhy-2 benchmark. The project page is available at https://tom-zgt.github.io/VPT.

cs.CV↗
arXiv2026

Nested Simulation Methods for Sobol' Index Estimation: Bias Correction, Budget Allocation, and Latin Hypercube Sampling

Estimating the variance of a conditional expectation is a recurring problem in stochastic simulation, with applications in global sensitivity analysis and Sobol' index estimation. This paper revisits Sobol' index estimation through the lens of nested simulation and develops a unified comparison of classical pick-freeze estimators and nested simulation estimators under a common computational budget. We show that several standard pick-freeze estimators can be interpreted as nested simulation estimators with fixed inner-level sample sizes, enabling direct performance comparisons and clarifying their bias-variance behavior. Building on this perspective, we analyze the standard nested simulation estimator for the Sobol' index numerator and propose two jackknife-based extensions: an unbiased jackknife estimator and a split jackknife estimator that uses an independent preliminary sample to estimate the mean. Under crude Monte Carlo (CMC), the split jackknife estimator attains the canonical mean squared error (MSE) rate, whereas the standard nested simulation and unbiased jackknife estimators attain the slower nested simulation rate. We also characterize the associated allocations of outer- and inner-level simulation effort. Finally, we study the impact of Latin hypercube sampling (LHS), showing that it can improve the standard nested simulation estimator while undermining bias reduction in jackknife-based estimators unless the inner-level sample size grows with the total budget. Numerical experiments corroborate the theory and provide practical guidance on estimator selection for Sobol' index estimation under CMC and LHS.

stat.ME↗
arXiv2026

Compact Spin-Charge Separated Neural Quantum States for Valence-Bond States

Neural-network quantum states (NQS) provide a flexible nonlinear representation of quantum many-body wavefunctions, but their efficiency depends sensitively on whether the architecture reflects the sign structure and constrained Hilbert space of the target state. In this work, we propose a solvable-point-guided strategy: design the architecture at an exactly solvable point where the correct local rules can be read off, then refine to the non-exact regime by enlarging only the kernel size and hidden dimension. The strategy is built from four physics-motivated designs: a stride-matched local-rule convolution, geometric pooling, a sign-resolving $\tanh(x^{2k+1})$ activation, and explicit spin-hole sector separation. We test this approach on quasi-one-dimensional valence-bond-solid (VBS) states and their doped soliton variants (sVBS), the exact ground states of a $t$-$J$-like model with a single mobile hole. In finite-size benchmarks, this architecture reaches high fidelity for the exact sVBS state with substantially fewer parameters than generic fully connected, convolutional, and transformer baselines tested under the same setup. For the spin sector, the learned local rule transfers from small to larger systems without retraining. Away from the solvable point, increasing kernel size and hidden dimension systematically improves accuracy, and the model shows approximately $L^2$ parameter scaling in the gapless regime for system size $L$, compared with approximately $L^4$ for matrix-product states in the same regime. Our work establishes a recipe for compact NQS in sign-structured, constrained Hilbert spaces and paves the pathway to physics-informed architectures for the broader $t$-$J$ and Hubbard families.

cond-mat.str-el↗
arXiv2026

Modeling Quantum Geometry for Fractional Chern Insulators with unsupervised learning

Fractional Chern insulators (FCIs) in moire materials present a unique platform for exploring strongly correlated topological phases beyond the paradigm of ideal quantum geometry. While analytical approaches to FCIs and fractional quantum Hall states (FQHS) often rely on idealized Bloch wavefunctions, realistic moire models lack direct tunability of quantum metric and Berry curvature, limiting theoretical and numerical exploration. Here, we introduce an unsupervised machine learning framework to model interacting Hamiltonians directly through the distribution of single-particle form factors. Using a variational autoencoder (VAE), we show that unsupervised learning can not only distinguish FCI and non-FCI states, but also generate new form factors with distinct topological character, not present in the training set. This latent space enables the generation and interpolation of form factors for topological flatbands with Chern number $|C|=1$, enabling the discovery of unobserved many-body states such as charge density waves. Principal component analysis (PCA) further reveals that the dominant patterns in the form factors-reflecting correlations across the Brillouin zone-can be decomposed into components with approximately quantized Chern numbers, providing new insights into the global and topological structure of quantum geometry. Our results highlight the ability of machine learning to generalize and model topological quantum systems, paving the way for the inverse design of form factors with tailored quantum geometry and many-body phases in flatband materials.

cond-mat.str-el↗
arXiv2026

SWIFT: Prompt-Adaptive Memory for Efficient Interactive Long Video Generation

Streaming long-video generation faces a central challenge in continuous semantic switching, requiring adaptive memory to preserve coherent visual evolution. Current approaches rely on cache rebuilding at prompt boundaries or fixed memory budgets, but they introduce redundant computation and limit flexible semantic adaptation. This limitation arises from a mismatch between cached video history and prompt updates, as memory preserves visual continuity while prompt switches demand rapid semantic adaptation. Motivated by this observation, we present SWIFT, Semantic Windowing and Injection for Flexible Transitions, a training-free framework for multi-prompt long-video generation that enables efficient semantic switching while preserving temporal coherence in causal video diffusion models. SWIFT introduces a lightweight Semantic Injection Cache that augments cached video memory rather than reconstructing it from scratch at every prompt boundary. To avoid uniformly perturbing all attention channels, we further perform head-wise semantic injection, so that each attention head receives a prompt update proportional to its alignment with the current video state. In addition, we introduce an Adaptive Dynamic Window that allocates temporal memory according to prompt phase, using larger local context near switching boundaries and smaller windows during stable segments to reduce average inference cost. To preserve long-range semantic consistency under compressed local attention, we further maintain segment-level semantic anchors that summarize prompt-conditioned video history and reintroduce it as compact memory tokens. Compared with current state-of-the-art methods, SWIFT preserves generation quality while achieving 22.6 FPS on a single H100 GPU, establishing a substantially more efficient solution for multi-prompt long-video generation. Our code is available at https://github.com/ShanwenTan/SWIFT.

cs.CV↗
arXiv2026

Adaptive Multi-Scale Channel-Spatial Attention Aggregation Framework for 3D Indoor Semantic Scene Completion Toward Assisting Visually Impaired

Independent indoor mobility remains a critical challenge for individuals with visual impairments, largely due to the limited capability of existing assistive systems in detecting fine-grained hazardous objects such as chairs, tables, and small obstacles. These perceptual blind zones substantially increase the risk of collision in unfamiliar environments. To bridge the gap between monocular 3D vision research and practical assistive deployment, this paper proposes an Adaptive Multi-scale Attention Aggregation (AMAA) framework for monocular 3D semantic scene completion using only a wearable RGB camera. The proposed framework addresses two major limitations in 2D-to-3D feature lifting: noise diffusion during back-projection and structural instability in multi-scale fusion. A parallel channel--spatial attention mechanism is introduced to recalibrate lifted features along semantic and geometric dimensions, while a hierarchical adaptive gating strategy regulates cross-scale information flow to preserve fine-grained structural details. Experiments on the NYUv2 benchmark demonstrate that AMAA achieves an overall mIoU of 27.88%. Crucially, it yields significant relative improvements of 16.9% for small objects and 10.4% for tables over the MonoScene baseline. Furthermore, a wearable prototype based on an NVIDIA Jetson Orin NX and a ZED~2i camera validates stable real-time performance in indoor environments, demonstrating the feasibility of deploying monocular 3D scene completion for assistive navigation.

cs.CV↗
arXiv2026

Trajectory Generation with Endpoint Regulation and Momentum-Aware Dynamics for Visually Impaired Scenarios

Trajectory generation for visually impaired scenarios requires smooth and temporally consistent state in structured, low-speed dynamic environments. However, traditional jerk-based heuristic trajectory sampling with independent segment generation and conventional smoothness penalties often lead to unstable terminal behavior and state discontinuities under frequent regenerating. This paper proposes a trajectory generation approach that integrates endpoint regulation to stabilize terminal states within each segment and momentum-aware dynamics to regularize the evolution of velocity and acceleration for segment consistency. Endpoint regulation is incorporated into trajectory sampling to stabilize terminal behavior, while a momentum-aware dynamics enforces consistent velocity and acceleration evolution across consecutive trajectory segments. Experimental results demonstrate reduced acceleration peaks and lower jerk levels with decreased dispersion, smoother velocity and acceleration profiles, more stable endpoint distributions, and fewer infeasible trajectory candidates compared with a baseline planner.

cs.RO↗
arXiv2026

RPIQ: Residual-Projected Multi-Collaboration Closed-Loop and Single Instance Quantization for Visually Impaired Assistance

Visually impaired users face significant challenges in daily information access and real-time environmental perception, and there is an urgent need for intelligent assistive systems with accurate recognition capabilities. Although large-scale models provide effective solutions for perception and reasoning, their practical deployment on assistive devices is severely constrained by excessive memory consumption and high inference costs. Moreover, existing quantization strategies often ignore inter-block error accumulation, leading to degraded model stability. To address these challenges, this study proposes a novel quantization framework -- Residual-Projected Multi-Collaboration Closed-Loop and Single Instance Quantization(RPIQ), whose quantization process adopts a multi-collaborative closed-loop compensation scheme based on Single Instance Calibration and Gauss-Seidel Iterative Quantization. Experiments on various types of large-scale models, including language models such as OPT, Qwen, and LLaMA, as well as vision-language models such as CogVLM2, demonstrate that RPIQ can compress models to 4-bit representation while significantly reducing peak memory consumption (approximately 60%-75% reduction compared to original full-precision models). The method maintains performance highly close to full-precision models across multiple language and visual tasks, and exhibits excellent recognition and reasoning capabilities in key applications such as text understanding and visual question answering in complex scenarios. While verifying the effectiveness of RPIQ for deployment in real assistive systems, this study also advances the computational efficiency and reliability of large models, enabling them to provide visually impaired users with the required information accurately and rapidly.

cs.LG↗
arXiv2026

Scene-Aware Vectorized Memory Multi-Agent Framework with Cross-Modal Differentiated Quantization VLMs for Visually Impaired Assistance

Visually impaired individuals face significant challenges in environmental perception. Traditional assistive technologies often lack adaptive intelligence, focusing on individual components rather than integrated systems. While Vision-Language Models (VLMs) offer a promising path to richer, integrated understanding, their deployment is severely limited by substantial computational requirements, demanding dozens of gigabytes of memory. To address these gaps in computational efficiency and integrated design, this study proposes a dual technological innovation framework: a cross-modal differentiated quantization framework for VLMs and a scene-aware vectorized memory multi-agent system. The quantization framework implements differentiated strategies, reducing memory from 38GB to 11.3GB. The multi-agent system uses vectorized memory and perception-memory-reasoning workflows to provide environmental information beyond the current view, achieving 2.83-3.52s latency to initial speech output. Experiments show the quantized 19B-parameter model only experiences a 2.05% performance drop on MMBench and maintains 63.7 accuracy on OCR-VQA (original: 64.9), outperforming smaller models with equivalent memory. This research advances computational efficiency and assistive technology, offering comprehensive assistance in scene perception, text recognition, and navigation.

cs.CV↗
arXiv2025

Multilevel Monte Carlo Metamodeling for Variance Function Estimation

This work introduces a novel multilevel Monte Carlo (MLMC) metamodeling approach for variance function estimation. Although devising an efficient experimental design for simulation metamodeling can be elusive, the MLMC-based approach addresses this challenge by dynamically adjusting the number of design points and budget allocation at each level, thereby automatically creating an efficient design. Theoretical analyses show that, under mild conditions, the proposed MLMC metamodeling approach for variance function estimation can achieve superior computational efficiency compared to standard Monte Carlo metamodeling while achieving the desired level of accuracy. Additionally, this work establishes the asymptotic normality of the MLMC metamodeling estimator under certain sufficient conditions, providing valuable insights for uncertainty quantification. Finally, two MLMC metamodeling procedures are proposed for variance function estimation: one to achieve a target accuracy level and another to efficiently utilize a fixed computational budget. Numerical evaluations support the theoretical results and demonstrate the potential of the proposed approach in facilitating global sensitivity analysis.

stat.ME↗