arXiv subjects

Ziteng Wang

Publications and source records attributed to Ziteng Wang.

At least 37 records · Page 2Linked to original sources

arXiv2026

The ASKAP Variables and Slow Transients (VAST) Extragalactic Survey - Data Release 1

The Variables and Slow Transients (VAST) Survey on the Australian SKA Pathfinder (ASKAP) is designed to systematically explore the dynamic radio sky, detecting sources that vary on timescales from minutes to several years. In this paper, we present Data Release 1 of the VAST Extragalactic Survey, which targets slowly evolving synchrotron transients in the southern sky. The observations were carried out between June 2023 and May 2025, comprising 2945 images of 276 fields spanning $\sim 12300\ \mathrm{deg}^2$, observed at 888 MHz with a typical rms sensitivity of 0.24 mJy $\rm{beam}^{-1}$ and 12-20 arcsec resolution. Each field was revisited approximately every two months, yielding 10 or 11 observations per field. The VAST pipeline extracts the light curves for all the observed sources, and additional filters are implemented to improve the reliability of the resulting light curve database. The light curve database contains 0.5 million sources and 6.4 million individual measurements, publicly available through the CSIRO data access portal. An untargeted variability search yields 117 astrophysical variables, including 27 pulsars, 40 radio stars (10 newly detected at radio wavelengths), 44 active galactic nuclei, two optically identified supernovae, one supernova candidate, one brown dwarf, and two sources without multi-wavelength counterparts that are yet to be identified. This data release provides the first large-scale, high-cadence, uniform view of long-term radio variability in the extragalactic sky and lays the groundwork for future population studies of radio transients with ASKAP.

astro-ph.CO↗

arXiv2026

ASKAP J005512.2-255834: A Luminous, Long-Lived Radio Transient at z = 0.1 -- an Orphan Afterglow or an off-nuclear TDE from an IMBH?

We report the discovery of a slowly evolving, extragalactic radio transient, ASKAP J005512.2--255834 (hereafter ASKAP J0055-2558), identified using the Australian SKA Pathfinder in a search for orphan afterglows associated with archival gravitational wave events. Although discovered in this context, there is no evidence that the transient is associated with any known gravitational wave event. Nonetheless, this source exhibits a 20-fold increase in flux density over $<250$ days, and it remains in a declining, detectable state more than 1000 days after the initial detection. Follow-up observations from 0.3 to 9 GHz reveal an evolving spectrum consistent with synchrotron emission. ASKAP J0055-2558 is spatially coincident with a low-mass, star-forming galaxy at redshift $z = 0.116$ ($d_{\rm L}$= 543 Mpc), placing its peak radio luminosity at $νL_ν\sim 10^{39}\,\rm erg\,s^{-1}$. Analysis of its radio light curve, inferred blastwave velocity, energetics, host galaxy properties and the absence of counterparts at other wavelengths suggest that ASKAP J0055-2558 is most consistent with either the late-time phase of an orphan long gamma-ray burst afterglow or a tidal disruption event involving an intermediate-mass black hole spatially offset from the galaxy nucleus. The radio discovery of either of these phenomena is extremely rare, with only a few or no confirmed examples to date.

astro-ph.HE↗

arXiv2026

Topological photonics in one-dimensional settings

Over the past decade, topological photonics has emerged as a vibrant field, attracting significant attention and witnessing remarkable advancements. This growth can be attributed to its fundamental appeal and the unique opportunities it offers for unconventional control of light, promising innovations in next-generation photonic devices. At the heart of topological photonics lies the one-dimensional (1D) SSH model. Originally conceived to elucidate the physics of a molecular chain of polyacetylene, this model has found widespread applications in exploring a wide range of topological phenomena in photonics and beyond. In this chapter, we aim to provide an overview of topological photonics in one-dimensional (1D) settings. After briefly introducing paradigmatic 1D models, including the SSH, Rice-Mele, and AAH models, we review recent advances in experimental studies and applications of topological photonics based on 1D platforms. Our discussion highlights demonstrated examples, such as the nonlinear tuning of topological states in both Hermitian and non-Hermitian photonic SSH lattices, as well as nonlinear harmonic generation and topological lasing in SSH-type photonic microstructures. We further discuss characteristic topological phenomena in other representative 1D settings, including Floquet systems, topological pumping, quasicrystals, and synthetic non-Hermitian systems. Finally, we examine selected examples of two-dimensional (2D) photonic topological crystalline insulators that are closely linked to the SSH model. Towards the end, we summarize the chapter and provide a list of key contributions, together with an outlook on possible future directions in 1D topological photonics. While this review focuses specifically on 1D topological photonics, it is not intended to be comprehensive or exhaustive.

physics.optics↗

arXiv2026

Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders

Representation Autoencoders (RAEs) have shown distinct advantages in diffusion modeling on ImageNet by training in high-dimensional semantic latent spaces. In this work, we investigate whether this framework can scale to large-scale, freeform text-to-image (T2I) generation. We first scale RAE decoders on the frozen representation encoder (SigLIP-2) beyond ImageNet by training on web, synthetic, and text-rendering data, finding that while scale improves general fidelity, targeted data composition is essential for specific domains like text. We then rigorously stress-test the RAE design choices originally proposed for ImageNet. Our analysis reveals that scaling simplifies the framework: while dimension-dependent noise scheduling remains critical, architectural complexities such as wide diffusion heads and noise-augmented decoding offer negligible benefits at scale Building on this simplified framework, we conduct a controlled comparison of RAE against the state-of-the-art FLUX VAE across diffusion transformer scales from 0.5B to 9.8B parameters. RAEs consistently outperform VAEs during pretraining across all model scales. Further, during finetuning on high-quality datasets, VAE-based models catastrophically overfit after 64 epochs, while RAE models remain stable through 256 epochs and achieve consistently better performance. Across all experiments, RAE-based diffusion models demonstrate faster convergence and better generation quality, establishing RAEs as a simpler and stronger foundation than VAEs for large-scale T2I generation. Additionally, because both visual understanding and generation can operate in a shared representation space, the multimodal model can directly reason over generated latents, opening new possibilities for unified models.

cs.CV↗

arXiv2026

V-FAT: Benchmarking Visual Fidelity Against Text-bias

Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated impressive performance on standard visual reasoning benchmarks. However, there is growing concern that these models rely excessively on linguistic shortcuts rather than genuine visual grounding, a phenomenon we term Text Bias. In this paper, we investigate the fundamental tension between visual perception and linguistic priors. We decouple the sources of this bias into two dimensions: Internal Corpus Bias, stemming from statistical correlations in pretraining, and External Instruction Bias, arising from the alignment-induced tendency toward sycophancy. To quantify this effect, we introduce V-FAT (Visual Fidelity Against Text-bias), a diagnostic benchmark comprising 4,026 VQA instances across six semantic domains. V-FAT employs a Three-Level Evaluation Framework that systematically increases the conflict between visual evidence and textual information: (L1) internal bias from atypical images, (L2) external bias from misleading instructions, and (L3) synergistic bias where both coincide. We introduce the Visual Robustness Score (VRS), a metric designed to penalize "lucky" linguistic guesses and reward true visual fidelity. Our evaluation of 12 frontier MLLMs reveals that while models excel in existing benchmarks, they experience significant visual collapse under high linguistic dominance.

cs.CL↗

arXiv2026

Phase Transitions and Topological Protection in Anyonic-PT-Symmetric Lattices

Parity-time (PT) symmetry and anti-PT symmetry have attracted extensive interest for their non-Hermitian spectral properties, particularly the emergence of purely real and imaginary eigenvalues in their symmetry-unbroken regime, respectively. Recently, these two scenarios have been unified under a more general framework known as anyonic-PT symmetry, yet its physical implications in waveguide platforms and corresponding topological features in extended lattice systems remain largely unexplored. Here, the phase transitions and topological protection in anyonic-PT-symmetric systems are systematically investigated in waveguide lattices. In the symmetry-unbroken regime, the arguments of all bulk eigenvalues are constrained to two discrete values separated by π, leading to distinctive oscillatory propagation dynamics accompanied by controlled amplification or dissipation. In the case of one-dimensional lattice, the energy bands exhibit a gap closing and reopening during phase transition. Moreover, in the symmetry-unbroken regime, the topological edge states emerge within the bulk gap and are protected by a generalized pseudo-anyonic-Hermiticity (PAH) symmetry. Our results establish anyonic-PT symmetry as a new tunable degree of freedom for non-Hermitian waveguide systems, where the eigenvalue argument provides a natural quantity for information encoding. This work broadens the conceptual foundation of topological protection under generalized non-Hermitian symmetries.

physics.optics↗

arXiv2026

Subsymmetry-protected compact edge states

Sub-symmetry (SubSy) protected topological states represent a concept that goes beyond the conventional framework of symmetry-protected topological (SPT) phases, demonstrating that topological boundary states can remain robust even when the pertinent symmetry holds only in a subset of Hilbert space. Typical SPT and SubSy boundary states decay exponentially into the bulk, which means they are not confined in just few lattice sites close to the boundary. Here, we introduce topologically compact edge states protected by SubSy, featuring extreme two-site localization at boundaries of a lattice, without any decay into the bulk. The compactness arises from local destructive interference at the boundary, while topological protection is ensured by SubSy, characterized by quantized winding numbers. Experimentally, we observe compact edge states in laser-written photonic lattices with engineered rhombic-like unit cells, confirming their robustness against perturbations under both chiral symmetry and SubSy conditions. Our results highlight the potential of SubSy protection for achieving topological confinement of light, paving the way for applications in compact waveguides, lasers, and high-sensitivity photonic sensors.

physics.optics↗

arXiv2025

JoyVoice: Long-Context Conditioning for Anthropomorphic Multi-Speaker Conversational Synthesis

Large speech generation models are evolving from single-speaker, short sentence synthesis to multi-speaker, long conversation geneartion. Current long-form speech generation models are predominately constrained to dyadic, turn-based interactions. To address this, we introduce JoyVoice, a novel anthropomorphic foundation model designed for flexible, boundary-free synthesis of up to eight speakers. Unlike conventional cascaded systems, JoyVoice employs a unified E2E-Transformer-DiT architecture that utilizes autoregressive hidden representations directly for diffusion inputs, enabling holistic end-to-end optimization. We further propose a MM-Tokenizer operating at a low bitrate of 12.5 Hz, which integrates multitask semantic and MMSE losses to effectively model both semantic and acoustic information. Additionally, the model incorporates robust text front-end processing via large-scale data perturbation. Experiments show that JoyVoice achieves state-of-the-art results in multilingual generation (Chinese, English, Japanese, Korean) and zero-shot voice cloning. JoyVoice achieves top-tier results on both the Seed-TTS-Eval Benchmark and multi-speaker long-form conversational voice cloning tasks, demonstrating superior audio quality and generalization. It achieves significant improvements in prosodic continuity for long-form speech, rhythm richness in multi-speaker conversations, paralinguistic naturalness, besides superior intelligibility. We encourage readers to listen to the demo at https://jea-speech.github.io/JoyVoice

cs.SD↗

arXiv2025

The radio afterglow of the ultra-long GRB 220627A

We present the discovery of the radio afterglow of the most distant ultra-long gamma-ray burst (GRB) detected to date, GRB~220627A at redshift $z=3.084$. Its prompt gamma-ray light curve shows a double-pulse profile, with the pulses separated by a period of quiescence lasting ${\sim} 15$\,min, leading to early speculation it could be a strongly gravitationally lensed GRB. However, our analysis of the \textit{Fermi}/GBM spectra taken during the time intervals of both pulses show clear differences in their spectral energy distributions, disfavouring the lensing scenario. We observed the radio afterglow from 7 to 456\,d post-burst: an initial, steep decay ($F_ν \propto t^{-2}$) is followed by a shallower decline ($F_ν \propto t^{-1/2}$) after ${\sim} 20$\,d. There are three scenarios that could explain these radio properties: (i) energy injection from an additional, slower ejecta component catching up to the external shock; (ii) a stratified density profile going as $n \propto r^{-8/3}$; or alternatively, (iii) the presence of a slow, wide ejecta component in addition to a fast, narrow ejecta component. We also conducted an independent test of the lensing hypothesis via Very Long Baseline Interferometry (VLBI) observations at ${\sim} 12$\,d post-burst by searching, for the first time, for multiple images of the candidate lensed GRB afterglow. Our experiment highlighted the growing need for developments in real-time correlation capabilities for time-critical VLBI experiments, particularly as we advance towards the SKA and ngVLA era of radio astronomy.

astro-ph.HE↗

arXiv2025

SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse-Linear Attention

In Diffusion Transformer (DiT) models, particularly for video generation, attention latency is a major bottleneck due to the long sequence length and the quadratic complexity. We find that attention weights can be separated into two parts: a small fraction of large weights with high rank and the remaining weights with very low rank. This naturally suggests applying sparse acceleration to the first part and low-rank acceleration to the second. Based on this finding, we propose SLA (Sparse-Linear Attention), a trainable attention method that fuses sparse and linear attention to accelerate diffusion models. SLA classifies attention weights into critical, marginal, and negligible categories, applying O(N^2) attention to critical weights, O(N) attention to marginal weights, and skipping negligible ones. SLA combines these computations into a single GPU kernel and supports both forward and backward passes. With only a few fine-tuning steps using SLA, DiT models achieve a 20x reduction in attention computation, resulting in significant acceleration without loss of generation quality. Experiments show that SLA reduces attention computation by 95% without degrading end-to-end generation quality, outperforming baseline methods. In addition, we implement an efficient GPU kernel for SLA, which yields a 13.7x speedup in attention computation and a 2.2x end-to-end speedup in video generation on Wan2.1-1.3B. The code is available at https://github.com/thu-ml/SLA.

cs.LG↗

arXiv2025

The Distribution of Atomic Hydrogen in the Host Galaxies of FRBs

We probe the atomic hydrogen (HI) emission from the host galaxies of fast radio bursts (FRBs) to investigate the emerging trend of disturbance and asymmetry in the population. Quadrupling the sample size, we detect 16 out of 17 new hosts in HI, with the single non-detection arising in a galaxy known to be transitioning towards quiescence. With respect to typical local Universe galaxies, FRB hosts are generally massive in HI ($M_{HI}>10^9 M_\odot$), which aligns with previous studies reporting that FRB hosts also tend to have high stellar masses and are star-forming. However, they span a broad range of other HI derived properties. Using visual inspection alongside various asymmetry metrics, we identify six unambiguously settled host galaxies, demonstrating for the first time that a disturbed HI morphology is not a universal feature of FRB host galaxies. However, we find another six that show clear signs of disturbance, one borderline case, and three which require deeper or more targeted observations to reach a conclusion; this brings the confirmed ratio of disturbed-to-settled FRB hosts to 11:6. Given that roughly a 1:1 ratio is expected for random background galaxies of similar type, our observed ratio yields a p-value of 0.222. Therefore, we conclude that contrary to earlier indications, there is no statistically significant excess of HI disturbance in this sample of FRB host galaxies with respect to the general galaxy population, and hence we find no evidence for a fundamental connection between FRB progenitor formation and merger-induced star formation activity.

astro-ph.GA↗

arXiv2025

VITRIX-CLIPIN: Enhancing Fine-Grained Visual Understanding in CLIP via Instruction Editing Data and Long Captions

Despite the success of Vision-Language Models (VLMs) like CLIP in aligning vision and language, their proficiency in detailed, fine-grained visual comprehension remains a key challenge. We present CLIP-IN, a novel framework that bolsters CLIP's fine-grained perception through two core innovations. Firstly, we leverage instruction-editing datasets, originally designed for image manipulation, as a unique source of hard negative image-text pairs. Coupled with a symmetric hard negative contrastive loss, this enables the model to effectively distinguish subtle visual-semantic differences. Secondly, CLIP-IN incorporates long descriptive captions, utilizing rotary positional encodings to capture rich semantic context often missed by standard CLIP. Our experiments demonstrate that CLIP-IN achieves substantial gains on the MMVP benchmark and various fine-grained visual recognition tasks, without compromising robust zero-shot performance on broader classification and retrieval tasks. Critically, integrating CLIP-IN's visual representations into Multimodal Large Language Models significantly reduces visual hallucinations and enhances reasoning abilities. This work underscores the considerable potential of synergizing targeted, instruction-based contrastive learning with comprehensive descriptive information to elevate the fine-grained understanding of VLMs.

cs.CV↗

arXiv2025

CRUST-Bench: A Comprehensive Benchmark for C-to-safe-Rust Transpilation

C-to-Rust transpilation is essential for modernizing legacy C code while enhancing safety and interoperability with modern Rust ecosystems. However, no dataset currently exists for evaluating whether a system can transpile C into safe Rust that passes a set of test cases. We introduce CRUST-Bench, a dataset of 100 C repositories, each paired with manually-written interfaces in safe Rust as well as test cases that can be used to validate correctness of the transpilation. By considering entire repositories rather than isolated functions, CRUST-Bench captures the challenges of translating complex projects with dependencies across multiple files. The provided Rust interfaces provide explicit specifications that ensure adherence to idiomatic, memory-safe Rust patterns, while the accompanying test cases enforce functional correctness. We evaluate state-of-the-art large language models (LLMs) on this task and find that safe and idiomatic Rust generation is still a challenging problem for various state-of-the-art methods and techniques. We also provide insights into the errors LLMs usually make in transpiling code from C to safe Rust. The best performing model, OpenAI o1, is able to solve only 15 tasks in a single-shot setting. Improvements on CRUST-Bench would lead to improved transpilation systems that can reason about complex scenarios and help in migrating legacy codebases from C into languages like Rust that ensure memory safety. You can find the dataset and code at https://github.com/anirudhkhatry/CRUST-bench.

cs.SE↗

arXiv2025

Mapping the Spatial Distribution of Fast Radio Bursts within their Host Galaxies

We present deep optical and near-infrared observations of the host galaxies of 34 fast radio bursts (FRBs) detected by the Commensal Real-time ASKAP Fast Transient (CRAFT) survey on the Australian SKA Pathfinder (ASKAP) to compare the locations of FRBs relative to their host light distributions. Incorporating three additional FRBs from the literature, for a total of four repeating and 33 apparently non-repeating FRBs, we determine their projected galactocentric offsets and find a median of $ 4.2^{+5.7}_{-2.5}$ kpc ($1.0^{+1.5}_{-0.6}r_e$). We model their host surface brightness profiles and develop synthetic spatial distributions of their globular clusters based on host properties. We calculate the likelihood the observed location of each FRB is consistent with the smooth light of its host galaxy, residual (primarily spiral) substructure, or globular cluster distributions. The majority of FRBs favor locations within the disks of their galaxies, while only 11$\pm$5\% favor a globular cluster origin, primarily those with galactocentric offsets $\gtrsim3r_e$. At $z<0.15$, where spiral structure is apparent in 86\% of our sample of FRB hosts, we find $\approx 20-46\%$ of FRBs favor an association with spiral arms. Assuming FRBs derive from magnetars, our results support multiple formation channels with the majority of progenitors associated with massive stars and a minority formed through dynamical channels. However, the moderate fraction of FRBs associated with spiral structure indicates that high star formation efficiency of the youngest and most massive stars is not a predominant driver in the production of FRB progenitors.

astro-ph.GA↗

arXiv2025

Quantum implicit representation of vortex filaments in turbulence

Entangled vortex filaments are essential to turbulence, serving as coherent structures that govern nonlinear fluid dynamics and support the reconstruction of fluid fields to reveal statistical properties. This study introduces an quantum implicit representation of vortex filaments in turbulence, employing a level-set method that models the filaments as the intersection of the real and imaginary zero iso-surfaces of a complex scalar field. Describing the fluid field via the scalar field offers distinct advantages in capturing complex structures, topological properties, and fluid dynamics, while opening new avenues for innovative solutions through quantum computing platforms. The representation is reformulated into an eigenvalue problem for Hermitian matrices, enabling the conversion of velocity fields into complex scalar fields that embed the vortex filaments. The resulting optimization is addressed using a variational quantum eigensolver, with Pauli operator truncation and deep learning techniques applied to improve efficiency and reduce noise. The proposed quantum framework achieves a near-linear time complexity and a exponential storage reduction while maintaining a balance of accuracy, robustness, and versatility, presenting a promising tool for turbulence analysis, vortex dynamics research, and machine learning dataset generation.

physics.flu-dyn↗

arXiv2025

A multi-ansatz variational quantum solver for compressible flows

Simulating nonlinear partial differential equations (PDEs) such as the Navier--Stokes (NS) equations remains computationally intensive, especially when implicit time integration is used to capture multiscale flow dynamics. This work introduces a hybrid quantum--classical framework for solving the linear systems arising from such implicit schemes in compressible flow simulations. At its core is a variational quantum linear solver (VQLS) enhanced by a multi-ansatz tree architecture, designed to expand the accessible solution space and alleviate training issues such as barren plateaus. The proposed method is evaluated through one-dimensional shock tube simulations implemented on a quantum virtual machine. Results demonstrate that the solver accurately captures shock, rarefaction, and contact discontinuities across a range of test cases. Parametric studies further show that increasing the number of ansatz branches and applying domain decomposition improves convergence and stability, even under limited qubit resources. These findings suggest that multi-ansatz VQLS architectures offer a promising pathway for incorporating quantum computing into computational fluid dynamics (CFD), with compatibility for both current noisy intermediate-scale quantum (NISQ) hardware and future fault-tolerant devices.

physics.flu-dyn↗

arXiv2025

Homomorphism Calculus for User-Defined Aggregations

Data processing frameworks like Apache Spark and Flink provide built-in support for user-defined aggregation functions (UDAFs), enabling the integration of domain-specific logic. However, for these frameworks to support \emph{efficient} UDAF execution, the function needs to satisfy a \emph{homomorphism property}, which ensures that partial results from independent computations can be merged correctly. Motivated by this problem, this paper introduces a novel \emph{homomorphism calculus} that can both verify and refute whether a UDAF is a dataframe homomorphism. If so, our calculus also enables the construction of a corresponding merge operator which can be used for incremental computation and parallel execution. We have implemented an algorithm based on our proposed calculus and evaluate it on real-world UDAFs, demonstrating that our approach significantly outperforms two leading synthesizers.

cs.PL↗

arXiv2025

A new long period radio transient: Discovery of pulses repeating every 1.16 hours from ASKAP J175534.9-252749.1

We report the discovery of several new pulses from the source ASKAP J175534.9-252749.1 (J1755-2527), originally identified from a single 2-min long pulse, confirming it as a long period transient (LPT) with a period of ~1.16 hours. The pulses are significantly scattered, consistent with Galactic electron density models. Two of the new pulses also had measurable polarisation, but unlike the originally detected pulse, the polarisation angle does not behave as expected from the rotating vector model. We interpret historical non-detections of J1755-2527 as an intrinsic intermittency that occurs on month-long timescales, and discuss possible causes. We conjecture that, like some other LPTs with periods >~ 1 hour, J1755-2527 may host a white dwarf in a binary orbit, but note that its period is marginally shorter than the canonical orbital period minimum of cataclysmic variables. Our work highlights the importance of additional observations in establishing the nature of unusual radio-emitting objects.

astro-ph.HE↗