arXiv subjects

Yu Zhang

Publications and source records attributed to Yu Zhang.

At least 487 records · Page 27Linked to original sources

arXiv2025

Object-level Correlation for Few-Shot Segmentation

Few-shot semantic segmentation (FSS) aims to segment objects of novel categories in the query images given only a few annotated support samples. Existing methods primarily build the image-level correlation between the support target object and the entire query image. However, this correlation contains the hard pixel noise, \textit{i.e.}, irrelevant background objects, that is intractable to trace and suppress, leading to the overfitting of the background. To address the limitation of this correlation, we imitate the biological vision process to identify novel objects in the object-level information. Target identification in the general objects is more valid than in the entire image, especially in the low-data regime. Inspired by this, we design an Object-level Correlation Network (OCNet) by establishing the object-level correlation between the support target object and query general objects, which is mainly composed of the General Object Mining Module (GOMM) and Correlation Construction Module (CCM). Specifically, GOMM constructs the query general object feature by learning saliency and high-level similarity cues, where the general objects include the irrelevant background objects and the target foreground object. Then, CCM establishes the object-level correlation by allocating the target prototypes to match the general object feature. The generated object-level correlation can mine the query target feature and suppress the hard pixel noise for the final prediction. Extensive experiments on PASCAL-${5}^{i}$ and COCO-${20}^{i}$ show that our model achieves the state-of-the-art performance.

cs.CV↗

arXiv2025

Thermodynamics and P-v criticality of RN-AdS black hole surrounded by PFDM on the EGUP framework

In this paper, we systematically study the thermodynamic properties and P-v criticality of the RN-AdS black hole surrounded by PFDM in the framework of the extended generalized uncertainty principle (EGUP). In the extended phase space, starting from the first law of black hole thermodynamics, the EGUP-corrected Hawking temperature, heat capacity, and entropy are derived. Then, we use graphical methods to analyze the effects of the EGUP parameters on these thermodynamic quantities. Analysis of the heat capacity indicates that the EGUP correction is beneficial for maintaining the thermodynamic stability of black holes. Finally, based on the modified thermodynamic quantities, we obtain the EGUP-corrected black hole equation of state and Gibbs free energy. The G-T diagram with the EGUP correction effect is analyzed, and the results show that when $P P_c$, we combine the analysis of the relationship between heat capacity and event horizon radius, finding that the phase transition characteristics depicted in the G-T diagram are similar to those without the correction effect.

gr-qc↗

arXiv2025

Presolving Convexified Optimal Power Flow with Mixtures of Gradient Experts

Convex relaxations and approximations of the optimal power flow (OPF) problem have gained significant research and industrial interest for planning and operations in electric power networks. One approach for reducing their solve times is presolving which eliminates constraints from the problem definition, thereby reducing the burden of the underlying optimization algorithm. To this end, we propose a presolving framework for convexified optimal power flow (C-OPF) problems, which uses a novel deep learning-based architecture called MoGE (Mixture of Gradient Experts). In this framework, problem size is reduced by learning the mapping between C-OPF parameters and optimal dual variables (the latter being representable as gradients), which is then used to screen constraints that are non-binding at optimum. The validity of using this presolve framework across arbitrary families of C-OPF problems is theoretically demonstrated. We characterize generalization in MoGE and develop a post-solve recovery procedure to mitigate possible constraint classification errors. Using two different C-OPF models, we show via simulations that our framework reduces solve times by upto 34% across multiple PGLIB and MATPOWER test cases, while providing an identical solution as the full problem.

math.OC↗

arXiv2025

Dynamical crossover between stretched- and compressed-exponential relaxation in a photoexcited crystal

Anomalous relaxation is one of the hallmarks of disordered systems. Following perturbation by an external source, many glassy, jammed and amorphous systems relax as a stretched or compressed exponential as a function of time. However, despite their ubiquity, the origins of and the connection between these phenomenological relaxation functions remains to be understood. Here, we observe a tunable crossover from stretched- to compressed-exponential relaxation by photoexciting single crystal Ca3Ru2O7 across a structural phase transition. We present a simple lattice model that shows how spatial inhomogeneity and local, strain-mediated interactions cooperate to produce the dynamical crossover. Our work reveals anomalous relaxation dynamics in an idealized single crystal material and establishes photoexcited solids as promising platforms for probing the mechanisms underlying anomalous relaxation.

cond-mat.str-el↗

arXiv2025

Efficient Entanglement Routing for Satellite-Aerial-Terrestrial Quantum Networks

In the era of 6G and beyond, space-aerial-terrestrial quantum networks (SATQNs) are shaping the future of the global-scale quantum Internet. This paper investigates the collaboration among satellite, aerial, and terrestrial quantum networks to efficiently transmit high-fidelity quantum entanglements over long distances. We begin with a comprehensive overview of existing satellite-, aerial-, and terrestrial-based quantum networks. Subsequently, we address the entanglement routing problem with the objective of maximizing quantum network throughput by jointly optimizing path selection and entanglement generation rates (PS-EGR). Given that the original problem is formulated as a mixed-integer linear programming (MILP) problem, which is inherently intractable, we propose a Benders' decomposition (BD)-based algorithm to solve the problem efficiently. Numerical results validate the effectiveness of the proposed PS-EGR scheme, offering valuable insights into various optimizable factors within the system. Finally, we discuss the current challenges and propose promising avenues for future research in SATQNs.

quant-ph↗

arXiv2025

From Experts to a Generalist: Toward General Whole-Body Control for Humanoid Robots

Achieving general agile whole-body control on humanoid robots remains a major challenge due to diverse motion demands and data conflicts. While existing frameworks excel in training single motion-specific policies, they struggle to generalize across highly varied behaviors due to conflicting control requirements and mismatched data distributions. In this work, we propose BumbleBee (BB), an expert-generalist learning framework that combines motion clustering and sim-to-real adaptation to overcome these challenges. BB first leverages an autoencoder-based clustering method to group behaviorally similar motions using motion features and motion descriptions. Expert policies are then trained within each cluster and refined with real-world data through iterative delta action modeling to bridge the sim-to-real gap. Finally, these experts are distilled into a unified generalist controller that preserves agility and robustness across all motion types. Experiments on two simulations and a real humanoid robot demonstrate that BB achieves state-of-the-art general whole-body control, setting a new benchmark for agile, robust, and generalizable humanoid performance in the real world. The project webpage is available at https://beingbeyond.github.io/BumbleBee/.

cs.RO↗

arXiv2025

OA-DET3D: Embedding Object Awareness as a General Plug-in for Multi-Camera 3D Object Detection

The recent advance in multi-camera 3D object detection is featured by bird's-eye view (BEV) representation or object queries. However, the ill-posed transformation from image-plane view to 3D space inevitably causes feature clutter and distortion, making the objects blur into the background. To this end, we explore how to incorporate supplementary cues for differentiating objects in the transformed feature representation. Formally, we introduce OA-DET3D, a general plug-in module that improves 3D object detection by bringing object awareness into a variety of existing 3D object detection pipelines. Specifically, OA-DET3D boosts the representation of objects by leveraging object-centric depth information and foreground pseudo points. First, we use object-level supervision from the properties of each 3D bounding box to guide the network in learning the depth distribution. Next, we select foreground pixels using a 2D object detector and project them into 3D space for pseudo-voxel feature encoding. Finally, the object-aware depth features and pseudo-voxel features are incorporated into the BEV representation or query feature from the baseline model with a deformable attention mechanism. We conduct extensive experiments on the nuScenes dataset and Argoverse 2 dataset to validate the merits of OA-DET3D. Our method achieves consistent improvements over the BEV-based baselines in terms of both average precision and comprehensive detection score.

cs.CV↗

arXiv2025

Conan: A Chunkwise Online Network for Zero-Shot Adaptive Voice Conversion

Zero-shot online voice conversion (VC) holds significant promise for real-time communications and entertainment. However, current VC models struggle to preserve semantic fidelity under real-time constraints, deliver natural-sounding conversions, and adapt effectively to unseen speaker characteristics. To address these challenges, we introduce Conan, a chunkwise online zero-shot voice conversion model that preserves the content of the source while matching the voice timbre and styles of reference speech. Conan comprises three core components: 1) a Stream Content Extractor that leverages Emformer for low-latency streaming content encoding; 2) an Adaptive Style Encoder that extracts fine-grained stylistic features from reference speech for enhanced style adaptation; 3) a Causal Shuffle Vocoder that implements a fully causal HiFiGAN using a pixel-shuffle mechanism. Experimental evaluations demonstrate that Conan outperforms baseline models in subjective and objective metrics. Audio samples can be found at https://aaronz345.github.io/ConanDemo.

eess.AS↗

arXiv2025

Estimated Informed Anytime Search for Sampling-Based Planning via Adaptive Sampler

Path planning in robotics often involves solving continuously valued, high-dimensional problems. Popular informed approaches include graph-based searches, such as A*, and sampling-based methods, such as Informed RRT*, which utilize informed set and anytime strategies to expedite path optimization incrementally. Informed sampling-based planners define informed sets as subsets of the problem domain based on the current best solution cost. However, when no solution is found, these planners re-sample and explore the entire configuration space, which is time-consuming and computationally expensive. This article introduces Multi-Informed Trees (MIT*), a novel planner that constructs estimated informed sets based on prior admissible solution costs before finding the initial solution, thereby accelerating the initial convergence rate. Moreover, MIT* employs an adaptive sampler that dynamically adjusts the sampling strategy based on the exploration process. Furthermore, MIT* utilizes length-related adaptive sparse collision checks to guide lazy reverse search. These features enhance path cost efficiency and computation times while ensuring high success rates in confined scenarios. Through a series of simulations and real-world experiments, it is confirmed that MIT* outperforms existing single-query, sampling-based planners for problems in R^4 to R^16 and has been successfully applied to real-world robot manipulation tasks. A video showcasing our experimental results is available at: https://youtu.be/30RsBIdexTU

cs.RO↗

arXiv2025

Deep Fuzzy Optimization for Batch-Size and Nearest Neighbors in Optimal Robot Motion Planning

Efficient motion planning algorithms are essential in robotics. Optimizing essential parameters, such as batch size and nearest neighbor selection in sampling-based methods, can enhance performance in the planning process. However, existing approaches often lack environmental adaptability. Inspired by the method of the deep fuzzy neural networks, this work introduces Learning-based Informed Trees (LIT*), a sampling-based deep fuzzy learning-based planner that dynamically adjusts batch size and nearest neighbor parameters to obstacle distributions in the configuration spaces. By encoding both global and local ratios via valid and invalid states, LIT* differentiates between obstacle-sparse and obstacle-dense regions, leading to lower-cost paths and reduced computation time. Experimental results in high-dimensional spaces demonstrate that LIT* achieves faster convergence and improved solution quality. It outperforms state-of-the-art single-query, sampling-based planners in environments ranging from R^8 to R^14 and is successfully validated on a dual-arm robot manipulation task. A video showcasing our experimental results is available at: https://youtu.be/NrNs9zebWWk

cs.RO↗

arXiv2025

DGenCTR: Towards a Universal Generative Paradigm for Click-Through Rate Prediction via Discrete Diffusion

Recent advances in generative models have inspired the field of recommender systems to explore generative approaches, but most existing research focuses on sequence generation, a paradigm ill-suited for click-through rate (CTR) prediction. CTR models critically depend on a large number of cross-features between the target item and the user to estimate the probability of clicking on the item, and discarding these cross-features will significantly impair model performance. Therefore, to harness the ability of generative models to understand data distributions and thereby alleviate the constraints of traditional discriminative models in label-scarce space, diverging from the item-generation paradigm of sequence generation methods, we propose a novel sample-level generation paradigm specifically designed for the CTR task: a two-stage Discrete Diffusion-Based Generative CTR training framework (DGenCTR). This two-stage framework comprises a diffusion-based generative pre-training stage and a CTR-targeted supervised fine-tuning stage for CTR. Finally, extensive offline experiments and online A/B testing conclusively validate the effectiveness of our framework.

cs.IR↗

arXiv2025

Elliptical K-Nearest Neighbors -- Path Optimization via Coulomb's Law and Invalid Vertices in C-space Obstacles

Path planning has long been an important and active research area in robotics. To address challenges in high-dimensional motion planning, this study introduces the Force Direction Informed Trees (FDIT*), a sampling-based planner designed to enhance speed and cost-effectiveness in pathfinding. FDIT* builds upon the state-of-the-art informed sampling planner, the Effort Informed Trees (EIT*), by capitalizing on often-overlooked information in invalid vertices. It incorporates principles of physical force, particularly Coulomb's law. This approach proposes the elliptical $k$-nearest neighbors search method, enabling fast convergence navigation and avoiding high solution cost or infeasible paths by exploring more problem-specific search-worthy areas. It demonstrates benefits in search efficiency and cost reduction, particularly in confined, high-dimensional environments. It can be viewed as an extension of nearest neighbors search techniques. Fusing invalid vertex data with physical dynamics facilitates force-direction-based search regions, resulting in an improved convergence rate to the optimum. FDIT* outperforms existing single-query, sampling-based planners on the tested problems in R^4 to R^16 and has been demonstrated on a real-world mobile manipulation task.

cs.RO↗

arXiv2025

Direction Informed Trees (DIT*): Optimal Path Planning via Direction Filter and Direction Cost Heuristic

Optimal path planning requires finding a series of feasible states from the starting point to the goal to optimize objectives. Popular path planning algorithms, such as Effort Informed Trees (EIT*), employ effort heuristics to guide the search. Effective heuristics are accurate and computationally efficient, but achieving both can be challenging due to their conflicting nature. This paper proposes Direction Informed Trees (DIT*), a sampling-based planner that focuses on optimizing the search direction for each edge, resulting in goal bias during exploration. We define edges as generalized vectors and integrate similarity indexes to establish a directional filter that selects the nearest neighbors and estimates direction costs. The estimated direction cost heuristics are utilized in edge evaluation. This strategy allows the exploration to share directional information efficiently. DIT* convergence faster than existing single-query, sampling-based planners on tested problems in R^4 to R^16 and has been demonstrated in real-world environments with various planning tasks. A video showcasing our experimental results is available at: https://youtu.be/2SX6QT2NOek

cs.RO↗

arXiv2025

UrgenGo: Urgency-Aware Transparent GPU Kernel Launching for Autonomous Driving

The rapid advancements in autonomous driving have introduced increasingly complex, real-time GPU-bound tasks critical for reliable vehicle operation. However, the proprietary nature of these autonomous systems and closed-source GPU drivers hinder fine-grained control over GPU executions, often resulting in missed deadlines that compromise vehicle performance. To address this, we present UrgenGo, a non-intrusive, urgency-aware GPU scheduling system that operates without access to application source code. UrgenGo implicitly prioritizes GPU executions through transparent kernel launch manipulation, employing task-level stream binding, delayed kernel launching, and batched kernel launch synchronization. We conducted extensive real-world evaluations in collaboration with a self-driving startup, developing 11 GPU-bound task chains for a realistic autonomous navigation application and implementing our system on a self-driving bus. Our results show a significant 61% reduction in the overall deadline miss ratio, compared to the state-of-the-art GPU scheduler that requires source code modifications.

cs.OS↗

arXiv2025

ExpertWeave: Efficiently Serving Expert-Specialized Fine-Tuned Adapters at Scale

Expert-Specialized Fine-Tuning (ESFT) adapts Mixture-of-Experts (MoE) large language models to enhance their task-specific performance by selectively tuning the top-activated experts for the task. Serving these fine-tuned models at scale is challenging: deploying merged models in isolation is prohibitively resource-hungry, while existing multi-adapter serving systems with LoRA-style additive updates are incompatible with ESFT's expert-oriented paradigm. We present ExpertWeave, a system that serves multiple ESFT adapters concurrently over a single shared MoE base model, drastically reducing the memory footprint and improving resource utilization. To seamlessly integrate into existing inference pipelines for MoE models with non-intrusive modifications and minimal latency overhead, ExpertWeave introduces a virtual-memory-assisted expert weight manager that co-locates base-model and adapter experts without incurring memory overhead from fragmentation, and a fused kernel for batched rerouting to enable lightweight redirection of tokens to the appropriate experts at runtime. Our evaluations show that ExpertWeave can simultaneously serve multiple adapters of a 16B MoE model on a single accelerator where the baseline runs out of memory, or provides up to 94x more KV cache capacity and achieves up to 18% higher throughput while using comparable resources, all without compromising model accuracy. ExpertWeave maintains low overhead even when scaling to 20 adapters, with a 4-11% latency increase compared with serving the base model alone. Source code will be released soon.

cs.DC↗

arXiv2025

InReAcTable: LLM-Powered Interactive Visual Data Story Construction from Tabular Data

Insights in tabular data capture valuable patterns that help analysts understand critical information. Organizing related insights into visual data stories is crucial for in-depth analysis. However, constructing such stories is challenging because of the complexity of the inherent relations between extracted insights. Users face difficulty sifting through a vast number of discrete insights to integrate specific ones into a unified narrative that meets their analytical goals. Existing methods either heavily rely on user expertise, making the process inefficient, or employ automated approaches that cannot fully capture their evolving goals. In this paper, we introduce InReAcTable, a framework that enhances visual data story construction by establishing both structural and semantic connections between data insights. Each user interaction triggers the Acting module, which utilizes an insight graph for structural filtering to narrow the search space, followed by the Reasoning module using the retrieval-augmented generation method based on large language models for semantic filtering, ultimately providing insight recommendations aligned with the user's analytical intent. Based on the InReAcTable framework, we develop an interactive prototype system that guides users to construct visual data stories aligned with their analytical requirements. We conducted a case study and a user experiment to demonstrate the utility and effectiveness of the InReAcTable framework and the prototype system for interactively building visual data stories.

cs.HC↗

arXiv2025

Exact Optimization for Minimum Dominating Sets

The Minimum Dominating Set (MDS) problem is a well-established combinatorial optimization problem with numerous real-world applications. Its NP-hard nature makes it increasingly difficult to obtain exact solutions as the graph size grows. This paper introduces ParDS, an exact algorithm developed to address the MDS problem within the branch-and-bound framework. ParDS features two key innovations: an advanced linear programming technique that yields tighter lower bounds and a set of novel reduction rules that dynamically simplify instances throughout the solving process. Compared to the leading exact algorithms presented at IJCAI 2023 and 2024, ParDS demonstrates theoretically superior lower-bound quality. Experimental results on standard benchmark datasets highlight several significant advantages of ParDS: it achieves fastest solving times in 70% of graph categories, especially on large, sparse graphs, delivers a speed-up of up to 3,411 times on the fastest individual instance, and successfully solves 16 out of 43 instances that other algorithms were unable to resolve within the 5-hour time limit. These findings establish ParDS as a state-of-the-art solution for exactly solving the MDS problem.

cs.DS↗

arXiv2025

Versatile Framework for Song Generation with Prompt-based Control

Song generation focuses on producing controllable high-quality songs based on various prompts. However, existing methods struggle to generate vocals and accompaniments with prompt-based control and proper alignment. Additionally, they fall short in supporting various tasks. To address these challenges, we introduce VersBand, a multi-task song generation framework for synthesizing high-quality, aligned songs with prompt-based control. VersBand comprises these primary models: 1) VocalBand, a decoupled model, leverages the flow-matching method for generating singing styles, pitches, and mel-spectrograms, allowing fast, high-quality vocal generation with style control. 2) AccompBand, a flow-based transformer model, incorporates the Band-MOE, selecting suitable experts for enhanced quality, alignment, and control. This model allows for generating controllable, high-quality accompaniments aligned with vocals. 3) Two generation models, LyricBand for lyrics and MelodyBand for melodies, contribute to the comprehensive multi-task song generation system, allowing for extensive control based on multiple prompts. Experimental results show that VersBand outperforms baseline models across multiple song generation tasks using objective and subjective metrics. Demos and codes are available at https://aaronz345.github.io/VersBandDemo and https://github.com/AaronZ345/VersBand.

eess.AS↗