Search arXiv⌕ Search

arXiv subjects

Yuqing Wang

Publications and source records attributed to Yuqing Wang.

At least 19 recordsLinked to original sources
arXiv2026

LESSViT: Robust Hyperspectral Representation Learning under Spectral Configuration Shift

Modeling hyperspectral imagery (HSI) across different sensors presents a fundamental challenge due to variations in wavelength coverage, band sampling, and channel dimensionality. As a result, models trained under a fixed spectral configuration often fail to generalize to other sensors. Existing Vision Transformer (ViT) approaches either rely on implicit spectral modeling with fixed channel assumptions or adopt explicit spatial-spectral attention with prohibitive computational cost, leading to a fundamental trade-off between efficiency and expressiveness. In this work, we introduce Low-rank Efficient Spatial-Spectral ViT (LESSViT), a sensor-flexible architecture for cross-spectral generalization. LESSViT is built on LESS Attention, a structured low-rank factorization that models joint spatial-spectral interactions through separable spatial and spectral components, reducing the complexity of full spatial-spectral attention from $\mathcal{O}(N^2 C^2)$ to $\mathcal{O}(rNC)$, where $N$ is the number of spatial tokens, $C$ is the number of spectral channels, and $r$ is the rank of the low-rank approximation. We further incorporate channel-agnostic patch embedding and wavelength-aware positional encoding to support flexible spectral inputs. To enable efficient and robust pretraining, we introduce a hyperspectral masked autoencoder (HyperMAE) with decoupled spatial-spectral masking and hierarchical channel sampling. We evaluate LESSViT under a cross-spectral generalization setting that simulates cross-sensor variability. Experiments on the SpectralEarth benchmark demonstrate that LESSViT improves robustness under spectral shifts while remaining competitive in-distribution, and explicit and efficient spatial-spectral modeling is essential for scalable and generalizable hyperspectral representation learning. The code is available through the project page: https://uiuctml.github.io/LESSViT/

cs.CV↗
arXiv2026

Fine-Grained Parallel Graph Algorithms Can Be Both Fast and Scalable

Making graph algorithms fast and scalable on traditional architectures and GPUs often involves breaking them into large, coarse-grained computational blocks that access similar regions of memory. We show that graph algorithms, which have tremendous vertex- and edge-level parallelism, can be scaled as fine-grained programs on suitable architectures. Specifically, we investigate graph algorithms on the UpDown architecture and show how this many-threaded, cache-less architecture enables simple, fine-grained parallel graph algorithms to be both fast and scalable. Single-node UpDown performance is 28x (geomean) over an H200 GPU, and that same program scales to 1000's of nodes. We evaluate the system three ways: simulation, analytical projection models, and selected validation studies at scale (simulation of 16,384 UpDown nodes, using 256 supercomputer nodes). These results show the UpDown system can achieve 680K giga-traversed edges per second (GTEPS) for PageRank and 989K GTEPS BFS, exceeding prior results significantly.

cs.DC↗
arXiv2026

Industrial Agentic AI: LLM-Driven New-Generation Agents for Smart Manufacturing

Large language models (LLMs) are enabling agents to move beyond predefined workflows toward autonomous planning and execution, creating new opportunities for smart manufacturing. However, dynamic environments, complex physical constraints, and stringent reliability requirements hinder the direct deployment of general-purpose agents in smart manufacturing. This article presents a systematic overview of LLM-driven industrial agents. First, we analyze five key challenges and discuss the concept and core characteristics of industrial agents. We then establish a unified taxonomy covering application scenarios, embodiment forms, and five capability levels. Next, we propose a four-layer system architecture and detail its core modules and operational mechanisms for closed-loop task execution. Finally, we review representative applications across the industrial lifecycle and outline future research directions. This article provides a structured reference for the design, development, and application of industrial agents.

cs.AI↗
arXiv2026

MomentBA: Second-order Spatial Moments for Anisotropic Correspondence Uncertainty in Differentiable Bundle Adjustment

Most existing visual odometry (VO) systems treat feature correspondences as deterministic measurements or assign uniform uncertainty, ignoring the inherent localization ambiguity of different observations. However, correspondence uncertainty is often anisotropic due to image structures such as edges, repetitive patterns, and motion blur, which can significantly affect geometric optimization. In this work, we propose MomentBA, a geometry-aware bundle adjustment framework that derives anisotropic correspondence uncertainty from second-order spatial moments of local similarity responses. Instead of introducing additional covariance prediction networks, the proposed method directly converts matching response distributions into interpretable covariance estimates and incorporates them into bundle adjustment as correspondence-specific information matrices for uncertainty-aware residual weighting. Furthermore, the proposed formulation is integrated into a differentiable optimization framework, establishing a direct connection between correspondence uncertainty and geometric estimation. Experiments on the EuRoC MAV and TartanAir v1 Hard datasets demonstrate that MomentBA improves monocular visual odometry accuracy compared with existing feature-based and learning-based approaches. The proposed anisotropic covariance model achieves lower rotational errors and more robust trajectory estimation than fixed and isotropic uncertainty models, validating the effectiveness of geometry-induced uncertainty modeling for challenging visual environments.

cs.CV↗
arXiv2026

A Theoretical Analysis of Generalization Dynamics in Neural Networks under Gradient Descent with Weight Decay

Understanding generalization remains a central challenge in machine learning because it requires jointly considering data, architecture, and training dynamics. In this paper, we develop a theoretical framework that characterizes how these factors jointly shape generalization performance throughout training. More precisely, we study a broad class of neural networks trained under the $\ell^2$ loss by gradient descent (GD) with weight decay, and prove the convergence of GD to a neighbourhood of the global minimizers of the empirical loss. By partitioning the space based on the input data, we then decompose the population error into data error, optimization error, and prediction variation error, and bound them separately. In particular, for the prediction variation error, which measures the oscillations of the learned function, we propose (local) approximate homogeneity and derive explicit cellwise and layerwise bounds for its evolution along the training trajectory. These bounds yield two important implications: a necessary condition of improved generalization explains differences in layerwise generalization behavior; a sufficient condition describes delayed generalization and provides a theoretical characterization of grokking.

cs.LG↗
arXiv2026

SARTM: Segment Any RGB Thermal Model with Language aided Distillation

The recent Segment Anything Model (SAM) demonstrates strong instance segmentation performance across various downstream tasks. However, SAM is trained solely on RGB data, limiting its direct applicability to RGB-thermal (RGB-T) semantic segmentation. Given that RGB-T provides a robust solution for scene understanding in adverse weather and lighting conditions, such as low light and overexposure, we propose a novel framework, SARTM, which customizes the powerful SAM for RGB-T semantic segmentation. Our key idea is to unleash the potential of SAM while introduce semantic understanding modules for RGB-T data pairs. Specifically, our framework first involves fine-tuning the original SAM by adding extra LoRA layers, aiming at preserving SAM's strong generalization and segmentation capabilities for downstream tasks. Secondly, we introduce language information as guidance for training our SARTM. To address cross-modal inconsistencies, we introduce a Cross-Modal Knowledge Distillation(CMKD) module that effectively achieves modality adaptation while maintaining its generalization capabilities. This semantic module enables the minimization of modality gaps and alleviates semantic ambiguity, facilitating the combination of any modality under any visual conditions. Furthermore, we enhance the segmentation performance by adjusting the segmentation head of SAM and incorporating an auxiliary semantic segmentation head, which integrates multi-scale features for effective fusion. Extensive experiments are conducted across three multi-modal RGBT semantic segmentation benchmarks: MFNET, PST900, and FMB. Both quantitative and qualitative results consistently demonstrate that the proposed SARTM significantly outperforms state-of-the-art approaches across a variety of conditions. Code and pre-trained weights can be found at https://github.com/wahaha-debug/SARTM.

cs.CV↗
arXiv2026

From Association to Causation: Improving Retrieval Precision of Retrieval-Augmented Generation via Causal Relations and an Attention Mechanism

Retrieval-Augmented Generation (RAG) grounds LLM generation on retrieved documents, but the standard terminal retrieval stage--dense-vector similarity, optionally followed by reranking--often returns documents that share keywords with the query without containing the needed information, a failure mode that grows with the knowledge base. We trace it to a conceptual gap: similarity captures only associational relations, whereas the documents that matter are linked to the query causally. We model the terminal retrieval stage with a causal graph grounded in Reichenbach's common cause principle: the keywords shared by the query and a retrieved document form a latent common cause A, and the document's residual keywords form a latent set B linking the document to the ideal output. Since a retrieved document is a collider (A -> d <- B), retrieval itself opens an associational path between the query and B, which licenses a training-free, attention-style re-scoring rule: the cosine similarity between the query embedding and the weighted centroid embedding of B. Unlike causality-enhanced RAG variants that model causal relations inside the knowledge content, our graph models the causal structure of the retrieval process itself. On a real 471-document enterprise knowledge base, the method promotes a relevant guideline from rank 6 to the top 3; on a controlled diagnostic corpus reproducing the keyword-stuffing regime, it improves the mean target rank from 2.88 to 1.25, while a trained cross-encoder reranker barely helps (2.63). Conversely, on three BEIR benchmarks the score underperforms the similarity baseline, delineating the applicability boundary: the method guards the keyword-stuffing regime of growing proprietary knowledge bases and complements neural rerankers; a corpus-level calibration gate selects the correct regime with >= 95% reliability. A fully local testbed demonstrates deployability.

cs.AI↗
arXiv2026

P2Fusion: Prompt-based Progressive Infrared-Visible Image Fusion via Dual-Prior Distillation

Infrared-visible image fusion (IVIF) is pivotal for multimodal perception, yet reconciling the inherent information disparity between thermal and textural features remains a fundamental challenge. Existing prior-guided methods often rely on static constraints that induce optimization conflicts or utilize extrinsic semantic priors from large-scale foundation models (e.g., CLIP/DINO), which frequently fail to exploit the intrinsic modality characteristics essential for high-fidelity fusion. To address these issues, we propose P2Fusion, a prior-guided distillation-based framework that reformulates IVIF via dual intrinsic prompts. Instead of imposing hard-coded penalties, we distill image-intrinsic priors, thermal saliency and spatial quality, into learnable dynamic regulators. Specifically, a Teach-to-Fuse mechanism provides dual-granularity progressive guidance, coupled with a Gated Dynamic Expert Recalibration (GDER) module for decoupled feature refinement. This design enables the network to adaptively mediate modal competition through expert specialization. Extensive experiments demonstrate that P2Fusion achieves state-of-the-art performance across five mainstream datasets. Notably, our framework demonstrates consistent performance advantages in fusion quality, achieving state-of-the-art results in 14 out of 20 key evaluation metrics across 5 benchmarks. Furthermore, it effectively contributes to the robustness of downstream perception, such as +3.2% mAP on MSRS, +0.5% mAP on M3FD and +0.9% mAP on DroneVehicle for object detection. Our code will be available at https://github.com/YiShi99/P2Fusion

cs.CV↗
arXiv2026

VADER: Adaptive Debiasing for Hallucination Mitigation in Video Large Language Models

Large vision-language models (LVLMs) have demonstrated strong performance in open-ended video understanding, yet they remain prone to fluent responses unsupported by video evidence. Existing training-free methods typically apply a globally fixed visual intervention or construct a contrastive branch through input perturbation. The former cannot accommodate video-dependent fusion paths, while the latter can be compensated by cross-frame redundancy. We therefore propose Video-Adaptive Debiasing via Evidence Reweighting (VADER), a training-free framework with two complementary modules. Visual Focus Reallocation (VFR) automatically instantiates an intervention policy for each video-question input: it diagnoses layer-wise visual-to-text evidence flow, determines where to intervene, and derives how strongly to reallocate pre-softmax attention from system-token to video-token blocks. Selective Evidence Erasure (SEE) independently masks high-importance visual tokens in every frame, constructing a prior-biased branch that is difficult to compensate through neighboring frames. Contrastive decoding then down-weights predictions that remain confident after selective evidence erasure. Across multiple VideoLLMs, VADER yields substantial improvements on event-level grounding and temporal consistency; on LLaVA-Video-7B, it reaches 72.60% accuracy on EventHallusion.

cs.CV↗
arXiv2026

ParticleGen: A Multi-Agent System for Particle Effects Generation

Particle systems are widely used in digital entertainment to create dynamic scene elements and visual effects. However, authoring high-quality particle effects remains labor-intensive and demands specialized expertise, requiring practitioners to navigate complex procedural rules and high-dimensional parameter spaces. Recent large language models (LLMs) enable users to specify particle effects through natural language, yet reliably translating high-level creative intent into executable procedural logic and low-level parameters remains difficult. In this work, we present a multi-agent framework for the from-scratch synthesis of structured and editable particle systems from natural language descriptions. Given a text prompt, our framework first generates an initial particle configuration through a decoupled planning and parameterization pipeline, and then iteratively improves the result based on rendered feedback. To support precise and targeted adjustments, we further introduce a diagnostic mechanism that links observed visual artifacts to their underlying procedural causes. We validate our approach in Unreal Engine 5's Niagara system across a diverse set of scenarios, including elemental spells, dynamic natural phenomena, and fireworks. Quantitative and qualitative evaluations show that our method achieves high semantic fidelity and visual quality. By directly synthesizing structured particle simulation logic, our framework reduces the technical barrier to particle effect authoring and improves the efficiency of creative iteration.

cs.GR↗
arXiv2026

i2Nav-Robot: A Large-Scale Indoor-Outdoor Robot Dataset for Multi-Sensor Fusion Navigation

Accurate and reliable navigation is crucial for autonomous unmanned ground vehicles (UGVs). However, current UGV datasets fall short in meeting the demands for advancing navigation techniques due to limitations in sensor configuration, time synchronization, ground truth, and scenario diversity. Hence, we present i2Nav-Robot, a large-scale dataset designed for multi-sensor fusion navigation in indoor-outdoor environments. We integrate multi-modal navigation sensors, including the newest front-view and 360-degree solid-state LiDARs, 4-dimensional (4D) millimeter-wave (MMW) radar, stereo cameras, inertial measurement units (IMU), global navigation satellite system (GNSS) receivers, and wheeled odometers on an omnidirectional wheeled vehicle. Accurate timestamps are obtained through both online hardware synchronization and offline calibration for all sensors. The dataset includes ten large-scale sequences covering diverse UGV operating scenarios, such as outdoor streets and indoor parking lots, with a total length of about 17060 meters. High-rate, reliable, and fully covered ground truth, with centimeter-level positioning, is derived from post-processing integrated navigation methods using a high-grade IMU. The proposed i2Nav-Robot dataset is evaluated by 15 open-sourced multi-sensor fusion navigation methods, demonstrating its superior data quality and utility for advancing vehicular navigation research. The i2Nav-Robot dataset together with the documents can be accessed on GitHub (https://github.com/i2Nav-WHU/i2Nav-Robot).

cs.RO↗
arXiv2026

Fault Injection in OpenAPI Specifications for Evaluating Black-Box Testing Effectiveness

OpenAPI specifications are the primary input for black-box testing tools in microservice systems (MSS), yet prior work shows these specifications are often incomplete, inconsistent, or incorrect. Despite this, most studies on OpenAPI-based black-box testing assume correct specifications and evaluate tool performance. We address this gap by introducing a literature-grounded taxonomy of six OpenAPI specification fault classes. We inject faults at five severity levels, and evaluate the resulting mutated specifications on two microservice benchmarks, TrainTicket and SocialNetwork, using three testing tools: EvoMaster, RESTler, and Schemathesis. We measure the impact of these faults using code coverage, specification coverage, request/response quality, and behavioral diversity. Our results show that specification faults cause strong and heterogeneous degradation patterns across testing tools and systems. Faults in method semantics cause broad degradation across all metrics, while others, such as modifications to response codes, remain weak. Relaxations of schema constraints cause hidden degradation, with no impact on code and specification coverage but a large impact on request/response quality. These findings demonstrate that specification quality directly shapes black-box API testing effectiveness. Also, code and specification coverage-only evaluations can understate the impact of specification faults on black-box testing in MSS and should be complemented by request/response quality and behavioral diversity.

cs.SE↗
arXiv2026

UniClawBench: A Universal Benchmark for Proactive Agents on Real-World Tasks

The rapid development of large language models and multimodal large language models has accelerated the emergence of proactive agents capable of operating everyday tools and assisting users in real-world environments. However, existing benchmarks struggle to evaluate such agents effectively, as they often rely on sandboxed environments and single-turn evaluation paradigms. Moreover, their scenario-based task taxonomies mix multiple model capabilities within the same task category, making it difficult to identify the root causes of agent failures. To address these limitations, we introduce UniClawBench, the first capability-driven benchmark designed to evaluate proactive agents in dynamic, real-world settings. UniClawBench is built around five foundational model capabilities: Skill Usage, Exploration, Long-Context Reasoning, Multimodal Understanding, and Cross-Platform Coordination. Based on these capabilities, we design 400 bilingual real-world tasks. Unlike previous benchmarks that rely on static, pre-recorded answers, our benchmark evaluates agents in live Docker containers using fine-grained, step-by-step completion checkpoints. Furthermore, we design a closed-loop evaluation strategy comprising an executor agent, a hidden supervisor agent, and a user agent to simulate realistic multi-turn human feedback without leaking grading criteria. To disentangle base model capabilities from framework-level design choices, we evaluate state-of-the-art models under multiple agent frameworks. Through comprehensive comparisons across both models and frameworks, we show how base model capabilities and agent framework designs jointly shape performance in real-world environments. To facilitate future research, we make our benchmark and code publicly available at https://github.com/HKU-MMLab/UniClawBench.

cs.CL↗
arXiv2026

Representation Forcing for Bottleneck-Free Unified Multimodal Models

Unified multimodal models (UMMs) aim to handle perception and generation in a single model. Yet existing UMMs still rely on a frozen, separately pretrained VAE for image generation, imposing a structural bottleneck. Naively removing it introduces a quality gap, as the model must learn both high-level structure and low-level details from raw pixels. In this paper, we propose Representation Forcing (RF), a technique that closes this gap by making representation prediction a native capability of the model. Concretely, RF forces the decoder to autoregressively predict visual representations as intermediate tokens before pixels; these tokens then stay in context to guide pixel diffusion within the same backbone. By turning representations from perception outputs into generation targets, RF eliminates the need for any external generative latent space. We find that RF benefits both understanding and generation. On image generation, our pixel-space model with RF matches state-of-the-art VAE-based unified models. On image understanding, pixel-space RF generally outperforms its VAE-based variant. Together, these results offer an effective step toward end-to-end, bottleneck-free UMMs.

cs.CV↗
arXiv2026

OpFlow: Learning Opportunity-Conditioned Choice Potentials for Robust OD Flow Prediction

Origin-destination (OD) flow prediction is central to urban analytics, yet deep models trained on raw counts remain vulnerable to distribution shift. The core problem is that raw count supervision cannot distinguish transferable choice mechanisms from environment-specific shortcuts. Raw OD count mixes two objects: how much demand an origin produces and how that demand is allocated across destinations. We argue that the transferable object is the exposure-to-choice law that maps spatial conditions to relative destination preferences. We propose OpFlow, a mechanism-constrained framework that learns row-centered choice potentials and reconstructs flows by combining the induced allocation with a separately calibrated origin scale. Under distribution shift, spatial exposures and the induced allocations are allowed to vary; what transfers is the conditional map from exposure states to relative choice potentials. Theoretically, we characterize the identifiable row-centered potential and show that classical spatial interaction laws are restricted log-potential cases. Controlled synthetic shifts and a real-world experiment show OpFlow improves robustness under environment shifts.

cs.LG↗
arXiv2026

Large Language Models for Fuzz Testing in Microservices: A Systematic Literature Review

Microservice systems (MSS) increasingly rely on heterogeneous APIs whose combinatorial input space and stateful dependencies challenge traditional fuzz testing. Meanwhile, Large Language Models (LLMs) have recently been introduced to enhance fuzzing with semantic reasoning over specifications, inputs, and runtime feedback. This paper presents a systematic literature review (SLR) of LLM-assisted fuzz testing for microservices to synthesise how LLMs are applied, evaluated, and what challenges remain. Following established SLR guidelines, we analyze 20 primary studies published between 2024 and 2026. Results show LLMs are mainly used as semantic input generators in black-box fuzzing, with a growing shift towards agent-based and retrieval-augmented architectures, improving valid input generation and modestly increasing coverage and vulnerability detection. However, evaluation remains heterogeneous, with limited benchmark standardization, scarce cost reporting, and a bias toward single-service experiments, highlighting a gap with real-world multi-service systems. This review provides a taxonomy of LLM roles and integrations, a consolidated view of evaluation practices, and a mapping of open challenges to research directions, supporting the design and deployment of LLM-driven fuzzing in microservices.

cs.SE↗
arXiv2026

Convergence of Gradient Descent for General Neural Network Architectures Beyond the NTK Regime

Training dynamics is central to understanding neural networks, yet its theoretical analysis remains difficult even for simple architectures and becomes substantially more challenging for general modern architectures. In this paper, we propose a convergence framework for analyzing gradient descent (GD) dynamics under a broad family of neural network architectures and datasets beyond the neural tangent kernel (NTK) regime. The framework is formulated at the level of network blocks and covers architectures including pre-normalized multi-layer transformers. More precisely, under mild assumptions, we prove that for almost all initializations, GD with regular learning rates converges to the neighbourhood of a stationary point. This is mainly proved by establishing an iterate-dependent PL-type inequality through analyticity and measure-zero arguments, and by proving Lipschitz smoothness along the GD trajectory through polynomial generalized smoothness and a local relaxed dissipative condition. We further interpret the theorem under Xavier initialization and practical architectural scaling, showing that the learning rate scale depends on the depth and effective bottleneck dimensions rather than the largest width. Finally, we derive structural nondegeneracy implications for residual connections and function composition, and provide a generic characterization of global minimizers within our framework.

cs.LG↗
arXiv2026

Towards LLM Accelerated Rapid Reviews for Software Tool Discovery -- Case for Log Anomaly Detection

In software engineering research, the primary outcome is frequently a tool. However, for practitioners and academics alike, it is hard to tell which tools are maintained and do they work out of the box. In this paper, we propose a pipeline to identify relevant studies with LLM screening, extract the tools presented in them, and run them with LLM-based coding agent. To evaluate the feasibility of our approach we focus on software log anomaly detection tools. We begin the study by designing a broad search string that yields 3233 hits from Scopus. We request two LLMs to provide an inclusion probability for each title-abstract pair according to the inclusion and exclusion criteria. From the 3233 exported abstracts, this screening reduced the number of included papers to 569, out of which we could download 470. These papers included 206 unique links and after manual evaluation we determined 83 to be tools. Finally, we ran the LLM-based coding agent on these 83 links, and got 24 successfully running tools. As replicating our approach would require roughly only 4 hours of human effort, of which 3 hours were manual PDF downloading, and 12 hours of LLM running time, this demonstrates promising efficiency when utilizing LLMs in rapid reviews. Because practitioner-built tools often lack academic papers, in the future we aim to expand our analysis to tool-hosting platforms such as GitHub and PyPI. In the future, we plan to formalize our workflow as LLM Agent Skills to make our approach easier to adopt.

cs.SE↗