arXiv subjects

Zijian Zhang

Publications and source records attributed to Zijian Zhang.

At least 109 records · Page 6Linked to original sources

arXiv2025

G2rammar: Bilingual Grammar Modeling for Enhanced Text-attributed Graph Learning

Text-attributed graphs require models to effectively integrate both structural topology and semantic content. Recent approaches apply large language models to graphs by linearizing structures into token sequences through random walks. These methods create concise graph vocabularies to replace verbose natural language descriptions. However, they overlook a critical component that makes language expressive: grammar. In natural language, grammar assigns syntactic roles to words and defines their functions within sentences. Similarly, nodes in graphs play distinct structural roles as hubs, bridges, or peripheral members. Current graph language methods provide tokens without grammatical annotations to indicate these structural or semantic roles. This absence limits language models' ability to reason about graph topology effectively. We propose \textbf{G2rammar}, a bilingual grammar framework that explicitly encodes both structural and semantic grammar for text-attributed graphs. Structural grammar characterizes topological roles through centrality and neighborhood patterns. Semantic grammar captures content relationships through textual informativity. The framework implements two-stage learning with structural grammar pre-training followed by semantic grammar fine-tuning. Extensive experiments on real-world datasets demonstrate that G2rammar consistently outperforms competitive baselines by providing language models with the grammatical context needed to understand graph structures.

cs.GR↗

arXiv2025

Reputation-Based Leader Election under Partial Synchrony: Towards a Protocol-Independent Abstraction with Enhanced Guarantees

Leader election serves a well-defined role in leader-based Byzantine Fault Tolerant (BFT) protocols. Existing reputation-based leader election frameworks for partially synchronous BFTs suffer from either protocol-specific proofs, narrow applicability, or unbounded recovery after network stabilization, leaving an open problem. This paper presents a novel protocol-independent abstraction formalizing generic correctness properties and effectiveness guarantees for leader election under partial synchrony, enabling protocol-independent analysis and design. Building on this, we design the Sliding Window Leader Election (SWLE) mechanism. SWLE dynamically adjusts leader nominations via consensus-behavior-based reputation scores, enforcing Byzantine-cost amplification. We demonstrate SWLE introduces minimal extra overhead to the base protocol and prove it satisfies all abstraction properties and provides superior effectiveness. We show, with a 16-server deployment across 4 different regions in northern China, SWLE achieves up to 4.2x higher throughput, 75% lower latency and 27% Byzantine leader frequency compared to the state-of-the-art solution under common Byzantine faults, while maintaining efficiency in fault-free scenarios.

cs.DC↗

arXiv2025

Bridging Quasars and Little Red Dots: Insights into Broad-Line AGNs at $z=5-8$ from the First JWST COSMOS-3D Dataset

We report the discovery of 13 broad-line AGNs at $z = 5 - 8$ from the first 10% data of the JWST Cycle 3 Treasury Program COSMOS-3D. These AGNs are identified by their broad H$α$ or H$β$ emission lines through the NIRCam grism wide-field slitless spectroscopy. One object at $z = 7.646$ with broad H$β$ emission has an F444W magnitude of 23.6 mag, making it one of the brightest $z > 7.5$ broad-line AGNs yet known. Among the 13 AGNs, 10 objects have reddened optical continua with slopes $β_{\rm opt}>0$. The remaining three objects have their overall SEDs that resemble those of UV-luminous quasars at similar redshifts, but their $β_{\rm opt}$, though negative, are not as blue as those of unobscured quasars. We also obtain MIRI photometry at 7.7-18 $μ$m for two AGNs and place strong constraints on their rest-frame near-IR SED. We find no significant variability in the rest-frame UV by comparing the COSMOS-3D and COSMOS-Web F115W images taken apart by 60 days in the rest-frame. We compute the H$α$ luminosity functions (LFs) for the broad H$α$ emitters at $z \approx 5-6$ and find a potential redshift evolution when compared with that of the $z \approx 4-5$ sample. We also derive the H$β$ LF at $z\sim8$ for AGNs and galaxies by combining our sample with those from the literature. The broad H$β$ emitters in this work suggest a number density two orders of magnitude higher than that predicted by the quasar LF based on rest-frame UV-selected samples. As a preview, our work showcases the ability of the COSMOS-3D grism survey to provide a complete view of the properties, growth, and evolution of bright broad-line AGNs at $z>5$.

astro-ph.GA↗

arXiv2025

Little red dot variability over a century reveals black hole envelope via a giant Einstein cross

"Little red dots" (LRDs) represent a new population of astronomical objects uncovered by JWST whose nature remains debated. Although many LRDs are suspected as active galactic nuclei (AGN), they show little variability on days-years timescales. We report the discovery of two gravitationally lensed LRDs at redshift $\sim$4.3 behind the cluster RXCJ2211-0350, one of which (RX1) is quadruply imaged with time delays spanning $\sim$130 years. RX1 exhibits intrinsic color and brightness variations of up to 0.7 magnitude among its images. These changes are consistent with blackbody-temperature variations of a photosphere, indicating long-term variability analogous to Cepheid-like pulsations but in a far more extended ($R \sim 2000$ AU) and massive ($M \gtrsim 10^6 \, M_{\odot}$) systems. These results suggest LRDs as a distinct class of AGN with stellar-like envelopes.

astro-ph.GA↗

arXiv2025

Multi-Scenario Highway Lane-Change Intention Prediction: A Physics-Informed AI Framework for Three-Class Classification

Lane-change maneuvers are a leading cause of highway accidents, underscoring the need for accurate intention prediction to improve the safety and decision-making of autonomous driving systems. While prior studies using machine learning and deep learning methods (e.g., SVM, CNN, LSTM, Transformers) have shown promise, most approaches remain limited by binary classification, lack of scenario diversity, and degraded performance under longer prediction horizons. In this study, we propose a physics-informed AI framework that explicitly integrates vehicle kinematics, interaction feasibility, and traffic-safety metrics (e.g., distance headway, time headway, time-to-collision, closing gap time) into the learning process. lane-change prediction is formulated as a three-class problem that distinguishes left change, right change, and no change, and is evaluated across both straight highway segments (highD) and complex ramp scenarios (exiD). By integrating vehicle kinematics with interaction features, our machine learning models, particularly LightGBM, achieve state-of-the-art accuracy and strong generalization. Results show up to 99.8% accuracy and 93.6% macro F1 on highD, and 96.1% accuracy and 88.7% macro F1 on exiD at a 1-second horizon, outperforming a two-layer stacked LSTM baseline. These findings demonstrate the practical advantages of a physics-informed and feature-rich machine learning framework for real-time lane-change intention prediction in autonomous driving systems.

cs.AI↗

arXiv2025

Discovery of two little red dots transitioning into quasars

James Webb Space Telescope (JWST) has uncovered a new population of compact objects that show a unique V-shaped spectral energy distribution (SED) in the UV and optical wavelength range. These so-called "little red dots" (LRDs) often exhibit broad Balmer emission lines, indicative of the presence of active galactic nuclei (AGNs). They generally lack detection of X-ray, radio, and mid-IR radiation, which is fundamentally different from typical AGNs. Various models, including super-Eddington-accreting black holes enshrouded in dense and dust-poor gas, have been proposed to explain these features. However, the nature of LRDs remains debated, and their evolutionary fate is unclear. Here we report two unusual LRDs at redshift z = 2.868 and 2.925 that are in a transitional phase towards typical AGNs. Their V-shaped SEDs, compact optical morphology, and broad emission lines satisfy the defining criteria of LRDs. On the other hand, they exhibit intense X-ray, radio, and mid-IR radiation that is much stronger than previously known LRDs. These hybrid properties suggest that the dense gas envelope around their central black holes is dispersing, allowing high-energy photons and radio emission to escape. Meanwhile, a dust torus is forming. This finding provides a direct insight into the nature of LRDs and indicates that at least some LRDs will evolve into AGNs or quasars at later times.

astro-ph.GA↗

arXiv2025

NeutrSHINE: a high repetition rate ultrafast neutron source driven by SHINE electron beam

Neutrons serve as unique probes for exploring the microscopic structure of matter, with the performance of a neutron source fundamentally governing the depth of scientific exploration and the breadth of industrial applicability. To address application demands including nuclear data measurement in the ultra-high-energy region, fundamental particle physics research, highly efficient non-destructive neutron testing, and extreme environment simulation, an ultrafast neutron source driven by the 8 GeV electron beam from the Shanghai high-repetition-rate extreme light facility (SHINE) was conceptually proposed, named NeutrSHINE. Using multidisciplinary simulation tools, key neutronic parameters, thermal behavior of high-power neutron targets, and the factors affecting the time resolution of the source were analyzed. The results affirm the technical feasibility and promising application prospects of the NeutrSHINE concept.

physics.acc-ph↗

arXiv2025

An Efficient Transient Nonlinear Circuit Simulator Using Exponential Integration and Block-Jacobi Precondition

Transient simulation of linear and nonlinear circuits remains an important task in modern EDA tools. At present, SPICE-like simulators face challenges in parallelization, nonlinear convergence and linear efficiency, especially when applied to large-scale circuits. To address the limitations of simulators in handling various nonlinear circuits, we adopt a generalized row-echelon regularization approach, which extends the applicability of exponential integrators to a broader class of differential algebraic equations. The proposed method employs matrix exponential vector products to integrate the regularized system, allowing for a larger time step size while preserving accuracy and stability. Furthermore, in order to accelerate GMRES-based solvers within Newton-Raphson iterations, a structured block-Jacobi preconditioner is designed for linear systems. For locally coupled circuits, Additive Schwarz overlapping strategy is adopted to enhance the solution performance. Numerical experiments of various nonlinear circuit models show that under same hardware environment, our method achieves a speedup of 1.95$\times$-- 3.27$\times$ in total computation time compared to Backward Euler with Inexact Newton iterations, and time steps have decreased by an average of 60.70\% (up to 74.59\%). Compared with EI-NK method, total computing time of our method has a speedup of 1.08$\times$-- 1.79$\times$. These results highlight the potential of proposed method for scalable and nonlinear circuit simulation.

cs.CE↗

arXiv2025

Provably Minimum-Length Conformal Prediction Sets for Ordinal Classification

Ordinal classification has been widely applied in many high-stakes applications, e.g., medical imaging and diagnosis, where reliable uncertainty quantification (UQ) is essential for decision making. Conformal prediction (CP) is a general UQ framework that provides statistically valid guarantees, which is especially useful in practice. However, prior ordinal CP methods mainly focus on heuristic algorithms or restrictively require the underlying model to predict a unimodal distribution over ordinal labels. Consequently, they provide limited insight into coverage-efficiency trade-offs, or a model-agnostic and distribution-free nature favored by CP methods. To this end, we fill this gap by propose an ordinal-CP method that is model-agnostic and provides instance-level optimal prediction intervals. Specifically, we formulate conformal ordinal classification as a minimum-length covering problem at the instance level. To solve this problem, we develop a sliding-window algorithm that is optimal on each calibration data, with only a linear time complexity in K, the number of label candidates. The local optimality per instance further also improves predictive efficiency in expectation. Moreover, we propose a length-regularized variant that shrinks prediction set size while preserving coverage. Experiments on four benchmark datasets from diverse domains are conducted to demonstrate the significantly improved predictive efficiency of the proposed methods over baselines (by 15% decrease on average over four datasets).

cs.LG↗

arXiv2025

Breaking the Passive Learning Trap: An Active Perception Strategy for Human Motion Prediction

Forecasting 3D human motion is an important embodiment of fine-grained understanding and cognition of human behavior by artificial agents. Current approaches excessively rely on implicit network modeling of spatiotemporal relationships and motion characteristics, falling into the passive learning trap that results in redundant and monotonous 3D coordinate information acquisition while lacking actively guided explicit learning mechanisms. To overcome these issues, we propose an Active Perceptual Strategy (APS) for human motion prediction, leveraging quotient space representations to explicitly encode motion properties while introducing auxiliary learning objectives to strengthen spatio-temporal modeling. Specifically, we first design a data perception module that projects poses into the quotient space, decoupling motion geometry from coordinate redundancy. By jointly encoding tangent vectors and Grassmann projections, this module simultaneously achieves geometric dimension reduction, semantic decoupling, and dynamic constraint enforcement for effective motion pose characterization. Furthermore, we introduce a network perception module that actively learns spatio-temporal dependencies through restorative learning. This module deliberately masks specific joints or injects noise to construct auxiliary supervision signals. A dedicated auxiliary learning network is designed to actively adapt and learn from perturbed information. Notably, APS is model agnostic and can be integrated with different prediction models to enhance active perceptual. The experimental results demonstrate that our method achieves the new state-of-the-art, outperforming existing methods by large margins: 16.3% on H3.6M, 13.9% on CMU Mocap, and 10.1% on 3DPW.

cs.CV↗

arXiv2025

HyperD: Hybrid Periodicity Decoupling Framework for Traffic Forecasting

Accurate traffic forecasting plays a vital role in intelligent transportation systems, enabling applications such as congestion control, route planning, and urban mobility optimization. However, traffic forecasting remains challenging due to two key factors: (1) complex spatial dependencies arising from dynamic interactions between road segments and traffic sensors across the network, and (2) the coexistence of multi-scale periodic patterns (e.g., daily and weekly periodic patterns driven by human routines) with irregular fluctuations caused by unpredictable events (e.g., accidents, weather, or construction). To tackle these challenges, we propose HyperD (Hybrid Periodic Decoupling), a novel framework that decouples traffic data into periodic and residual components. The periodic component is handled by the Hybrid Periodic Representation Module, which extracts fine-grained daily and weekly patterns using learnable periodic embeddings and spatial-temporal attention. The residual component, which captures non-periodic, high-frequency fluctuations, is modeled by the Frequency-Aware Residual Representation Module, leveraging complex-valued MLP in frequency domain. To enforce semantic separation between the two components, we further introduce a Dual-View Alignment Loss, which aligns low-frequency information with the periodic branch and high-frequency information with the residual branch. Extensive experiments on four real-world traffic datasets demonstrate that HyperD achieves state-of-the-art prediction accuracy, while offering superior robustness under disturbances and improved computational efficiency compared to existing methods.

cs.AI↗

arXiv2025

Boosting Fine-Grained Urban Flow Inference via Lightweight Architecture and Focalized Optimization

Fine-grained urban flow inference is crucial for urban planning and intelligent transportation systems, enabling precise traffic management and resource allocation. However, the practical deployment of existing methods is hindered by two key challenges: the prohibitive computational cost of over-parameterized models and the suboptimal performance of conventional loss functions on the highly skewed distribution of urban flows. To address these challenges, we propose a unified solution that synergizes architectural efficiency with adaptive optimization. Specifically, we first introduce PLGF, a lightweight yet powerful architecture that employs a Progressive Local-Global Fusion strategy to effectively capture both fine-grained details and global contextual dependencies. Second, we propose DualFocal Loss, a novel function that integrates dual-space supervision with a difficulty-aware focusing mechanism, enabling the model to adaptively concentrate on hard-to-predict regions. Extensive experiments on 4 real-world scenarios validate the effectiveness and scalability of our method. Notably, while achieving state-of-the-art performance, PLGF reduces the model size by up to 97% compared to current high-performing methods. Furthermore, under comparable parameter budgets, our model yields an accuracy improvement of over 10% against strong baselines. The implementation is included in the https://github.com/Yasoz/PLGF.

cs.AI↗

arXiv2025

CudaForge: An Agent Framework with Hardware Feedback for CUDA Kernel Optimization

Developing efficient CUDA kernels is increasingly critical for AI applications such as large-scale LLM training. However, manual kernel design is both costly and time-consuming, motivating automatic approaches that leverage LLMs for code generation. Existing methods for automatic kernel generation, however, often produce low-efficiency kernels, incur high computational overhead, and fail to generalize across settings. In this work, we propose CudaForge, a training-free multi-agent workflow for CUDA kernel generation and optimization. Our workflow is inspired by the iterative workflow of human experts, which contains steps such as developing initial kernels, testing correctness, analyzing hardware feedback, and iterative improvement. More specifically, CudaForge employs two LLM agents: a Coder and a Judge, that iteratively generate, correct, and optimize CUDA kernels, while integrating hardware feedback such as Nsight Compute (NCU) metrics. In extensive evaluations, we show that CudaForge, by leveraging base models like OpenAI-o3, achieves 97.6\% correctness of generated kernels and an average 1.68$\times$ speedup over PyTorch baselines, substantially surpassing state-of-the-art models including OpenAI-o3 and Kevin on KernelBench.Beyond accuracy and speed, CudaForge demonstrates strong generalization across GPUs (A100, RTX 6000, 4090, 3090) and base models (OpenAI-o3, GPT-5, gpt-oss-120B, Claude-Sonnet-4, QwQ-32B), while maintaining high efficiency. In particular, generating an optimized kernel takes about 26.5 minutes on one RTX6000 and incurs about \$ 0.3 API cost, which is significantly cheaper than existing agentic work that costs 6 H100 hours and \$ 5 API cost per kernel. Our results highlight that multi-agent, training-free workflows can enable cost-effective, generalizable, and high-performance CUDA kernel optimization. Code available at https://github.com/OptimAI-Lab/CudaForge

cs.LG↗

arXiv2025

FTSmartAudit: A Knowledge Distillation-Enhanced Framework for Automated Smart Contract Auditing Using Fine-Tuned LLMs

The rapid growth of blockchain technology has driven the widespread adoption of smart contracts. However, their inherent vulnerabilities have led to significant financial losses. Traditional auditing methods, while essential, struggle to keep pace with the increasing complexity and scale of smart contracts. Large Language Models (LLMs) offer promising capabilities for automating vulnerability detection, but their adoption is often limited by high computational costs. Although prior work has explored leveraging large models through agents or workflows, relatively little attention has been given to improving the performance of smaller, fine-tuned models--a critical factor for achieving both efficiency and data privacy. In this paper, we introduce HKT-SmartAudit, a framework for developing lightweight models optimized for smart contract auditing. It features a multi-stage knowledge distillation pipeline that integrates classical distillation, external domain knowledge, and reward-guided learning to transfer high-quality insights from large teacher models. A single-task learning strategy is employed to train compact student models that maintain high accuracy and robustness while significantly reducing computational overhead. Experimental results show that our distilled models outperform both commercial tools and larger models in detecting complex vulnerabilities and logical flaws, offering a practical, secure, and scalable solution for smart contract auditing. The source code is available at Github repository.

cs.CR↗

arXiv2025

Decoupling What to Count and Where to See for Referring Expression Counting

Referring Expression Counting (REC) extends class-level object counting to the fine-grained subclass-level, aiming to enumerate objects matching a textual expression that specifies both the class and distinguishing attribute. A fundamental challenge, however, has been overlooked: annotation points are typically placed on class-representative locations (e.g., heads), forcing models to focus on class-level features while neglecting attribute information from other visual regions (e.g., legs for "walking"). To address this, we propose W2-Net, a novel framework that explicitly decouples the problem into "what to count" and "where to see" via a dual-query mechanism. Specifically, alongside the standard what-to-count (w2c) queries that localize the object, we introduce dedicated where-to-see (w2s) queries. The w2s queries are guided to seek and extract features from attribute-specific visual regions, enabling precise subclass discrimination. Furthermore, we introduce Subclass Separable Matching (SSM), a novel matching strategy that incorporates a repulsive force to enhance inter-subclass separability during label assignment. W2-Net significantly outperforms the state-of-the-art on the REC-8K dataset, reducing counting error by 22.5% (validation) and 18.0% (test), and improving localization F1 by 7% and 8%, respectively. Code will be available.

cs.CV↗

arXiv2025

Field-Free Superconducting Diode Enabled by Geometric Asymmetry and Perpendicular Magnetization

The superconducting diode effect (SDE)- manifested as directional, dissipationless supercurrents - is pivotal for realizing energy-efficient superconducting logic and memory technologies. Achieving high-efficiency SDE without external magnetic fields, however, remains a fundamental challenge. Here, we report a strongly enhanced, field-free SDE in Pt/Co/Nb heterostructures, enabled by the interplay of engineered geometric asymmetry and stray fields from a perpendicularly magnetized Co layer. This configuration promotes directional vortex entry and spatially selective pinning, yielding diode efficiencies that exceed all previously reported field-free values. Temperature- and field-dependent transport measurements, supported by micromagnetic simulations, reveal that the enhanced nonreciprocity stems from three cooperative mechanisms: asymmetric vortex entry, localized magnetic pinning, and Lorentz-force imbalance. These findings establish a scalable, CMOS-compatible platform for high-performance superconducting rectifiers, offering new opportunities for cryogenic spintronics and quantum electronics.

cond-mat.supr-con↗

arXiv2025

Discovery of a Little Red Dot candidate at $z\gtrsim10$ in COSMOS-Web based on MIRI-NIRCam selection

JWST has revealed a new high-redshift population called little red dots (LRDs). Since LRDs may be in the early phase of black hole growth, identifying them in the early universe is crucial for understanding the formation of the first supermassive black holes. However, no robust LRD candidates have been identified at $z>10$, because commonly-used NIRCam photometry covers wavelengths up to $\sim5\,{\rm μm}$ and is insufficient to capture the characteristic V-shaped spectral energy distributions (SEDs) of LRDs. In this study, we present the first search for $z\gtrsim10$ LRD candidates using both NIRCam and MIRI imaging from COSMOS-Web, which provides the largest joint NIRCam-MIRI coverage to date ($0.20\,{\rm deg^2}$). Taking advantage of MIRI/F770W to remove contaminants, we identify one robust candidate, CW-LRD-z10 at $z_{\rm phot}=10.5^{+0.7}_{-0.6}$ with $M_{\rm UV}=-19.9^{+0.1}_{-0.2}\,{\rm mag}$. CW-LRD-z10 exhibits a compact morphology, a distinct V-shaped SED, and a non-detection in F115W, all consistent with being an LRD at $z\sim10$. Based on this discovery, we place the first constraint on the number density of LRDs at $z\sim10$ with $M_{\rm UV}\sim-20$ of $1.2^{+2.7}_{-1.0}\times10^{-6}\,{\rm Mpc^{-3}\,mag^{-1}}$, suggesting that the fraction of LRDs among the overall galaxy population increases with redshift, reaching $\sim3\%$ at $z\sim10$. Although deep spectroscopy is necessary to confirm the redshift and the nature of CW-LRD-z10, our results imply that LRDs may be a common population at $z>10$, playing a key role in the first supermassive black hole formation.

astro-ph.GA↗

arXiv2025

TimeEmb: A Lightweight Static-Dynamic Disentanglement Framework for Time Series Forecasting

Temporal non-stationarity, the phenomenon that time series distributions change over time, poses fundamental challenges to reliable time series forecasting. Intuitively, the complex time series can be decomposed into two factors, \ie time-invariant and time-varying components, which indicate static and dynamic patterns, respectively. Nonetheless, existing methods often conflate the time-varying and time-invariant components, and jointly learn the combined long-term patterns and short-term fluctuations, leading to suboptimal performance facing distribution shifts. To address this issue, we initiatively propose a lightweight static-dynamic decomposition framework, TimeEmb, for time series forecasting. TimeEmb innovatively separates time series into two complementary components: (1) time-invariant component, captured by a novel global embedding module that learns persistent representations across time series, and (2) time-varying component, processed by an efficient frequency-domain filtering mechanism inspired by full-spectrum analysis in signal processing. Experiments on real-world datasets demonstrate that TimeEmb outperforms state-of-the-art baselines and requires fewer computational resources. We conduct comprehensive quantitative and qualitative analyses to verify the efficacy of static-dynamic disentanglement. This lightweight framework can also improve existing time-series forecasting methods with simple integration. To ease reproducibility, the code is available at https://github.com/showmeon/TimeEmb.

cs.LG↗