arXiv subjects

Zhen Zhang

Publications and source records attributed to Zhen Zhang.

At least 181 records · Page 10Linked to original sources

arXiv2025

Towards the Automatic Segmentation, Modeling and Meshing of the Aortic Vessel Tree from Multicenter Acquisitions: An Overview of the SEG.A. 2023 Segmentation of the Aorta Challenge

The automated analysis of the aortic vessel tree (AVT) from computed tomography angiography (CTA) holds immense clinical potential, but its development has been impeded by a lack of shared, high-quality data. We launched the SEG.A. challenge to catalyze progress in this field by introducing a large, publicly available, multi-institutional dataset for AVT segmentation. The challenge benchmarked automated algorithms on a hidden test set, with subsequent optional tasks in surface meshing for computational simulations. Our findings reveal a clear convergence on deep learning methodologies, with 3D U-Net architectures dominating the top submissions. A key result was that an ensemble of the highest-ranking algorithms significantly outperformed individual models, highlighting the benefits of model fusion. Performance was strongly linked to algorithmic design, particularly the use of customized post-processing steps, and the characteristics of the training data. This initiative not only establishes a new performance benchmark but also provides a lasting resource to drive future innovation toward robust, clinically translatable tools.

cs.CV↗

arXiv2025

AgentFrontier: Expanding the Capability Frontier of LLM Agents with ZPD-Guided Data Synthesis

Training large language model agents on tasks at the frontier of their capabilities is key to unlocking advanced reasoning. We introduce a data synthesis approach inspired by the educational theory of the Zone of Proximal Development (ZPD), which defines this frontier as tasks an LLM cannot solve alone but can master with guidance. To operationalize this, we present the AgentFrontier Engine, an automated pipeline that synthesizes high-quality, multidisciplinary data situated precisely within the LLM's ZPD. This engine supports both continued pre-training with knowledge-intensive data and targeted post-training on complex reasoning tasks. From the same framework, we derive the ZPD Exam, a dynamic and automated benchmark designed to evaluate agent capabilities on these frontier tasks. We train AgentFrontier-30B-A3B model on our synthesized data, which achieves state-of-the-art results on demanding benchmarks like Humanity's Last Exam, even surpassing some leading proprietary agents. Our work demonstrates that a ZPD-guided approach to data synthesis offers a scalable and effective path toward building more capable LLM agents.

cs.CL↗

arXiv2025

SharpZO: Hybrid Sharpness-Aware Vision Language Model Prompt Tuning via Forward-Only Passes

Fine-tuning vision language models (VLMs) has achieved remarkable performance across various downstream tasks; yet, it requires access to model gradients through backpropagation (BP), making them unsuitable for memory-constrained, inference-only edge devices. To address this limitation, previous work has explored various BP-free fine-tuning methods. However, these approaches often rely on high-variance evolutionary strategies (ES) or zeroth-order (ZO) optimization, and often fail to achieve satisfactory performance. In this paper, we propose a hybrid Sharpness-aware Zeroth-order optimization (SharpZO) approach, specifically designed to enhance the performance of ZO VLM fine-tuning via a sharpness-aware warm-up training. SharpZO features a two-stage optimization process: a sharpness-aware ES stage that globally explores and smooths the loss landscape to construct a strong initialization, followed by a fine-grained local search via sparse ZO optimization. The entire optimization relies solely on forward passes. Detailed theoretical analysis and extensive experiments on CLIP models demonstrate that SharpZO significantly improves accuracy and convergence speed, achieving up to 7% average gain over state-of-the-art forward-only methods.

cs.LG↗

arXiv2025

Presenting a Paper is an Art: Self-Improvement Aesthetic Agents for Academic Presentations

The promotion of academic papers has become an important means of enhancing research visibility. However, existing automated methods struggle limited storytelling, insufficient aesthetic quality, and constrained self-adjustment, making it difficult to achieve efficient and engaging dissemination. At the heart of those challenges is a simple principle: \emph{there is no way to improve it when you cannot evaluate it right}. To address this, we introduce \textbf{EvoPresent}, a self-improvement agent framework that unifies coherent narratives, aesthetic-aware designs, and realistic presentation delivery via virtual characters. Central to EvoPresent is \textbf{PresAesth}, a multi-task reinforcement learning (RL) aesthetic model that provides reliable aesthetic scoring, defect adjustment, and comparative feedback, enabling iterative self-improvement even under limited aesthetic training data. To systematically evaluate the methods, we introduce \textbf{EvoPresent Benchmark}, a comprehensive benchmark comprising: \textit{Presentation Generation Quality}, built on 650 top-tier AI conference papers with multimodal resources (slides, videos and scripts) to assess both content and design; and \textit{Aesthetic Awareness}, consisting of 2,000 slide pairs with varying aesthetic levels, supporting joint training and evaluation on scoring, defect adjustment, and comparison. Our findings highlight that (i) High-quality feedback is essential for agent self-improvement, while initial capability alone does not guarantee effective self-correction. (ii) Automated generation pipelines exhibit a trade-off between visual design and content construction. (iii) Multi-task RL training shows stronger generalization in aesthetic awareness tasks.

cs.CL↗

arXiv2025

Towards Unsupervised Open-Set Graph Domain Adaptation via Dual Reprogramming

Unsupervised Graph Domain Adaptation has become a promising paradigm for transferring knowledge from a fully labeled source graph to an unlabeled target graph. Existing graph domain adaptation models primarily focus on the closed-set setting, where the source and target domains share the same label spaces. However, this assumption might not be practical in the real-world scenarios, as the target domain might include classes that are not present in the source domain. In this paper, we investigate the problem of unsupervised open-set graph domain adaptation, where the goal is to not only correctly classify target nodes into the known classes, but also recognize previously unseen node types into the unknown class. Towards this end, we propose a novel framework called GraphRTA, which conducts reprogramming on both the graph and model sides. Specifically, we reprogram the graph by modifying target graph structure and node features, which facilitates better separation of known and unknown classes. Meanwhile, we also perform model reprogramming by pruning domain-specific parameters to reduce bias towards the source graph while preserving parameters that capture transferable patterns across graphs. Additionally, we extend the classifier with an extra dimension for the unknown class, thus eliminating the need of manually specified threshold in open-set recognition. Comprehensive experiments on several public datasets demonstrate that our proposed model can achieve satisfied performance compared with recent state-of-the-art baselines. Our source codes and datasets are publicly available at https://github.com/cszhangzhen/GraphRTA.

cs.LG↗

arXiv2025

Primal-dual splitting methods for phase-field surfactant model with moving contact lines

Surfactants have important effects on the dynamics of droplets on solid surfaces, which has inspired many industrial applications. Phase-field surfactant model with moving contact lines (PFS-MCL) has been employed to investigate the complex droplet dynamics with surfactants, while its numerical simulation remains challenging due to the coupling of gradient flows with respect to transport distances involving nonlinear and degenerate mobilities. We propose a novel structure-preserving variational scheme for PFS-MCL model with the dynamic boundary condition based on the minimizing movement scheme and optimal transport theory for Wasserstein gradient flows. The proposed scheme consists of a series of convex minimization problems and can be efficiently solved by our proposed primal-dual splitting method and its accelerated versions. By respecting the underlying PDE's variational structure with respect to the transport distance, the proposed scheme is proved to inherits the desirable properties including original energy dissipation, bound-preserving, and mass conservation. Through a suite of numerical simulations, we validate the performance of the proposed scheme and investigate the effects of surfactants on the droplet dynamics.

math.NA↗

arXiv2025

APEX: Empowering LLMs with Physics-Based Task Planning for Real-time Insight

Large Language Models (LLMs) demonstrate strong reasoning and task planning capabilities but remain fundamentally limited in physical interaction modeling. Existing approaches integrate perception via Vision-Language Models (VLMs) or adaptive decision-making through Reinforcement Learning (RL), but they fail to capture dynamic object interactions or require task-specific training, limiting their real-world applicability. We introduce APEX (Anticipatory Physics-Enhanced Execution), a framework that equips LLMs with physics-driven foresight for real-time task planning. APEX constructs structured graphs to identify and model the most relevant dynamic interactions in the environment, providing LLMs with explicit physical state updates. Simultaneously, APEX provides low-latency forward simulations of physically feasible actions, allowing LLMs to select optimal strategies based on predictive outcomes rather than static observations. We evaluate APEX on three benchmarks designed to assess perception, prediction, and decision-making: (1) Physics Reasoning Benchmark, testing causal inference and object motion prediction; (2) Tetris, evaluating whether physics-informed prediction enhances decision-making performance in long-horizon planning tasks; (3) Dynamic Obstacle Avoidance, assessing the immediate integration of perception and action feasibility analysis. APEX significantly outperforms standard LLMs and VLM-based models, demonstrating the necessity of explicit physics reasoning for bridging the gap between language-based intelligence and real-world task execution. The source code and experiment setup are publicly available at https://github.com/hwj20/APEX_EXP .

cs.RO↗

arXiv2025

Symmetry Transitions Beyond the Nanoscale in Pressurized Silica Glass

Silica is the paradigmatic network glass-former and understanding its response to pressure is essential for comprehending the mechanical properties of silica-based materials and the behavior of silicate melts in the Earth's interior. While pressure-induced changes in the short-range structure - particularly the breakdown of tetrahedral symmetry - have been well documented, structural transformations on larger length scales, important for many material properties, remain poorly understood. Here, we numerically investigate the three-dimensional structure of silica glass as a function of compression up to $P \approx 100$~GPa. Using a novel many-body correlation function, we reveal a complex medium-range order: While for $P \lesssim 10$~GPa, one finds tetrahedral, octahedral, and cubic symmetries, the structure at higher $P$s exhibits alternating cubic and octahedral particle arrangements. The $P$-dependence of the corresponding structural correlation length displays two distinct maxima, which permits to rationalize the anomalous compressibility of silica. The identified complex structural organization on intermediate range scales is the result of a pressure-and scale-dependent interplay between directional bonding, packing efficiency, and network stiffness. Since these competing effects are common in network glass-formers, the identified three-dimensional medium-range order, and hence the physical properties of the glass, are expected to be universal features of such materials under extreme conditions.

cond-mat.dis-nn↗

arXiv2025

Unveil A Peculiar Light Curve Pattern of Magnetar Burst with GECAM observations of SGR J1935+2154

Magnetar X-ray Burst (MXB) is usually composed of a single pulse or multiple pulses with rapid rise and brief duration mostly observed in hard X-ray (soft gamma-ray) band. Previous work studied the temporal behavior of some magnetar bursts and employed the Fast Rise Exponential Decay (FRED) model to fit pulses of MXB. However, whether there is other kind of pulse shape has not been explored. In this study, we systematically examined light curve of MXBs from SGR J1935+2154 detected by GECAM between 2021 and 2022. We find that there are different light curve morphologies. Especially, we discover a peculiar and new pattern, Exponential Rise and Cut-Off Decay (ERCOD), which is significantly different from FRED and could be well described by a mathematical function we proposed. We find that MXBs with ERCOD shape are generally longer in duration, brighter in the peak flux, and harder in spectrum. We note that the ERCOD shape is not unique to SGR J1935+2154 but also present in other magnetars. This new light curve pattern may imply a special burst and radiation mechanism of magnetar.

astro-ph.HE↗

arXiv2025

Forestpest-YOLO: A High-Performance Detection Framework for Small Forestry Pests

Detecting agricultural pests in complex forestry environments using remote sensing imagery is fundamental for ecological preservation, yet it is severely hampered by practical challenges. Targets are often minuscule, heavily occluded, and visually similar to the cluttered background, causing conventional object detection models to falter due to the loss of fine-grained features and an inability to handle extreme data imbalance. To overcome these obstacles, this paper introduces Forestpest-YOLO, a detection framework meticulously optimized for the nuances of forestry remote sensing. Building upon the YOLOv8 architecture, our framework introduces a synergistic trio of innovations. We first integrate a lossless downsampling module, SPD-Conv, to ensure that critical high-resolution details of small targets are preserved throughout the network. This is complemented by a novel cross-stage feature fusion block, CSPOK, which dynamically enhances multi-scale feature representation while suppressing background noise. Finally, we employ VarifocalLoss to refine the training objective, compelling the model to focus on high-quality and hard-to-classify samples. Extensive experiments on our challenging, self-constructed ForestPest dataset demonstrate that Forestpest-YOLO achieves state-of-the-art performance, showing marked improvements in detecting small, occluded pests and significantly outperforming established baseline models.

cs.CV↗

arXiv2025

A construction of simple-minded systems over domestic Brauer graph algebras I: the 2-domestic case

Let $A$ be a 2-domestic Brauer graph algebra. We present a construction for a family of objects on $A$-$\stmod$ to be a simple-minded system and our construction provides all simple-minded systems on $A$-$\stmod$. As a byproduct, we provide a new proof of AR-conjecture for 2-domestic Brauer graph algebras and we prove that a weakly simple-minded system with a finite cardinality is a simple-minded system on $A$-$\stmod$. We also prove that an orthogonal system $\mathcal{S}$ which contains at least one object for an Euclidean component extends to a simple-minded system on $A$-$\stmod$ and its extension closure $\mathcal{F}(\mathcal{S})$ is functorially finite.

math.RT↗

arXiv2025

Electric-field control of pure spin photocurrent in germanene

The electrical control of pure spin current remains a central challenge in spintronics, particularly in time-reversal symmetric systems composed of nonmagnetic elements, where spin and electric fields interact only indirectly. In this work, we develop a theoretical framework for electrically tuning pure spin photocurrent in two-dimensional materials with time-reversal symmetry via a gate electric field. Through theoretical analysis, we demonstrate that in systems with spin-orbit coupling and in-plane mirror symmetry, an out-of-plane electric field induces spin splitting and reversal in the band structure near the Fermi energy, enabling magnitude control and direction reversal of the pure spin photocurrent. To validate this mechanism, we perform first-principles calculations on germanene, an experimentally realized two-dimensional material. Beyond amplitude modulation, we reveal that reversing the direction of the applied electric field leads to a corresponding reversal of the pure spin photocurrent. Furthermore, we show that the pure spin photocurrent can be tuned by varying the photon energy and the incident angle of light, providing additional degrees of control over spin transport. These findings establish a robust strategy for electric-field-controlled pure spin transport in two-dimensional materials, offering new possibilities for the development of optospintronic devices.

cond-mat.mes-hall↗

arXiv2025

On the Value of Cross-Modal Misalignment in Multimodal Representation Learning

Multimodal representation learning, exemplified by multimodal contrastive learning (MMCL) using image-text pairs, aims to learn powerful representations by aligning cues across modalities. This approach relies on the core assumption that the exemplar image-text pairs constitute two representations of an identical concept. However, recent research has revealed that real-world datasets often exhibit cross-modal misalignment. There are two distinct viewpoints on how to address this issue: one suggests mitigating the misalignment, and the other leveraging it. We seek here to reconcile these seemingly opposing perspectives, and to provide a practical guide for practitioners. Using latent variable models we thus formalize cross-modal misalignment by introducing two specific mechanisms: Selection bias, where some semantic variables are absent in the text, and perturbation bias, where semantic variables are altered -- both leading to misalignment in data pairs. Our theoretical analysis demonstrates that, under mild assumptions, the representations learned by MMCL capture exactly the information related to the subset of the semantic variables invariant to selection and perturbation biases. This provides a unified perspective for understanding misalignment. Based on this, we further offer actionable insights into how misalignment should inform the design of real-world ML systems. We validate our theoretical findings via extensive empirical studies on both synthetic data and real image-text datasets, shedding light on the nuanced impact of cross-modal misalignment on multimodal representation learning.

cs.LG↗

arXiv2025

Scaling Laws for Online Advertisement Retrieval

The scaling law is a notable property of neural network models and has significantly propelled the development of large language models. Scaling laws hold great promise in guiding model design and resource allocation. Recent research increasingly shows that scaling laws are not limited to NLP tasks or Transformer architectures; they also apply to domains such as recommendation. However, there is still a lack of literature on scaling law research in online advertisement retrieval systems. This may be because 1) identifying the scaling law for resource cost and online revenue is often expensive in both time and training resources for industrial applications, and 2) varying settings for different systems prevent the scaling law from being applied across various scenarios. To address these issues, we propose a lightweight paradigm to identify online scaling laws of retrieval models, incorporating a novel offline metric and an offline simulation algorithm. We prove that under mild assumptions, the correlation between the novel metric and online revenue asymptotically approaches 1 and empirically validates its effectiveness. The simulation algorithm can estimate the machine cost offline. Based on the lightweight paradigm, we can identify online scaling laws for retrieval models almost exclusively through offline experiments, and quickly estimate machine costs and revenues for given model configurations. We further validate the existence of scaling laws across mainstream model architectures (e.g., Transformer, MLP, and DSSM) in our real-world advertising system. With the identified scaling laws, we demonstrate practical applications for ROI-constrained model designing and multi-scenario resource allocation in the online advertising system. To the best of our knowledge, this is the first work to study identification and application of online scaling laws for online advertisement retrieval.

cs.IR↗

arXiv2025

RuN: Residual Policy for Natural Humanoid Locomotion

Enabling humanoid robots to achieve natural and dynamic locomotion across a wide range of speeds, including smooth transitions from walking to running, presents a significant challenge. Existing deep reinforcement learning methods typically require the policy to directly track a reference motion, forcing a single policy to simultaneously learn motion imitation, velocity tracking, and stability maintenance. To address this, we introduce RuN, a novel decoupled residual learning framework. RuN decomposes the control task by pairing a pre-trained Conditional Motion Generator, which provides a kinematically natural motion prior, with a reinforcement learning policy that learns a lightweight residual correction to handle dynamical interactions. Experiments in simulation and reality on the Unitree G1 humanoid robot demonstrate that RuN achieves stable, natural gaits and smooth walk-run transitions across a broad velocity range (0-2.5 m/s), outperforming state-of-the-art methods in both training efficiency and final performance.

cs.RO↗

arXiv2025

WebResearcher: Unleashing unbounded reasoning capability in Long-Horizon Agents

Recent advances in deep-research systems have demonstrated the potential for AI agents to autonomously discover and synthesize knowledge from external sources. In this paper, we introduce WebResearcher, a novel framework for building such agents through two key components: (1) WebResearcher, an iterative deep-research paradigm that reformulates deep research as a Markov Decision Process, where agents periodically consolidate findings into evolving reports while maintaining focused workspaces, overcoming the context suffocation and noise contamination that plague existing mono-contextual approaches; and (2) WebFrontier, a scalable data synthesis engine that generates high-quality training data through tool-augmented complexity escalation, enabling systematic creation of research tasks that bridge the gap between passive knowledge recall and active knowledge construction. Notably, we find that the training data from our paradigm significantly enhances tool-use capabilities even for traditional mono-contextual methods. Furthermore, our paradigm naturally scales through parallel thinking, enabling concurrent multi-agent exploration for more comprehensive conclusions. Extensive experiments across 6 challenging benchmarks demonstrate that WebResearcher achieves state-of-the-art performance, even surpassing frontier proprietary systems.

cs.CL↗

arXiv2025

Prediction of Li3Fe8B8 compound with rapid one-dimensional ion diffusion channels

Using a computational crystal structure search in the Li-Fe-B ternary system, we predict a stable phase of Li3Fe8B8, featuring 1D channels that enable rapid Li-ion transport. Ab initio molecular dynamics simulations show that the Li-ion diffusion coefficient in Li3Fe8B8 surpasses that of common electrode and conductive additive materials by several orders of magnitude. The high diffusion in Li3Fe8B8 can be explained by the Frenkel-Kontorova model, which describes an incommensurate state between the Li diffusion chain and the periodic potential field caused by the FeB backbone structure. The favorable lithium-ion diffusivity and mechanical properties of Li3Fe8B8 make it a promising conductive additive for battery materials. Its itinerant ferromagnetism also offers a platform for exploring correlated-electron magnetism and spin-dependent phenomena.

cond-mat.mtrl-sci↗

arXiv2025

KBM: Delineating Knowledge Boundary for Adaptive Retrieval in Large Language Models

Large Language Models (LLMs) often struggle with dynamically changing knowledge and handling unknown static information. Retrieval-Augmented Generation (RAG) is employed to tackle these challenges and has a significant impact on improving LLM performance. In fact, we find that not all questions need to trigger RAG. By retrieving parts of knowledge unknown to the LLM and allowing the LLM to answer the rest, we can effectively reduce both time and computational costs. In our work, we propose a Knowledge Boundary Model (KBM) to express the known/unknown of a given question, and to determine whether a RAG needs to be triggered. Experiments conducted on 11 English and Chinese datasets illustrate that the KBM effectively delineates the knowledge boundary, significantly decreasing the proportion of retrievals required for optimal end-to-end performance. Furthermore, we evaluate the effectiveness of KBM in three complex scenarios: dynamic knowledge, long-tail static knowledge, and multi-hop problems, as well as its functionality as an external LLM plug-in.

cs.CL↗