arXiv subjects

Qian Zheng

Publications and source records attributed to Qian Zheng.

At least 19 recordsLinked to original sources

arXiv2026

Executable verification through formalized expert reasoning in astronomical spectroscopy

Artificial intelligence has reshaped scientific prediction, but scientific verification remains a human bottleneck. Automated systems can map observations to labels, parameters or hypotheses, yet scientific conclusions require evidence, must satisfy physical consistency, and need explicit testing of alternatives before a decision is made. Here we introduce FORMA (Formalized Observational Reasoning with Auditable Decisions), an executable verification protocol that reconstructs expert reasoning into a workflow: it extracts evidence, generates hypotheses under physical constraints, tests alternatives, and performs auditable consistency checks. Unlike prediction or post-hoc interpretability, executable verification records and tests the evidential path leading to a decision. Astronomical spectroscopy provides a natural testbed, because ambiguous survey spectra are still adjudicated by expert visual inspection. Applied to the Dark Energy Spectroscopic Instrument (DESI) visual inspection catalogue, FORMA combines template-fitting candidate redshifts, spectral evidence extraction and physical audit into an auditable credibility score. A medium-or-higher credibility threshold identifies $331$ definite predictions with $95.5\%$ binary agreement with expert-adjudicated classes, while increasing credibility is associated with improved redshift consistency and higher classification reliability. These results show that automated inference can be coupled to explicit verification, allowing candidate outputs to be evaluated before they enter scientific use.

astro-ph.CO↗

arXiv2026

Emergence of half-semiconductor behavior and tunable magnetism via carrier doping in Janus VXSe (X=Cl, Br, I) monolayers

Two-dimensional ferromagnetic semiconductors with high Curie temperature, large magnetic anisotropy, and electrically tunable properties are highly sought for nanoscale spintronics. Here, using first-principles calculations, we predict a new class of Janus VXSe (X=Cl, Br, I) monolayers. All three compounds are intrinsic ferromagnetic semiconductors with indirect band gaps of 1.66-2.33 eV, Curie temperatures up to 128 K, and magnetic anisotropy energies up to 910 ueV per V, leading to easy-plane magnetization for VClSe and VBrSe and an out-of-plane easy axis for VISe. The robust ferromagnetic order originates from the competition between superexchange and itinerant magnetism. Remarkably, VISe is a pristine half-semiconductor, whereas carrier doping unlocks fully spin-polarized states in the initially non-ideal VClSe and VBrSe. We also find that hole doping can switch the magnetic easy axis of VBrSe from in-plane to out-of-plane, enabling a spin-field-effect transistor with giant magnetoresistance. Our findings highlight carrier doping as a key to unlock hidden half-semiconducting behavior and establish Janus VXSe as a promising platform for 2D spintronics and magnetoelectric devices.

cond-mat.mtrl-sci↗

arXiv2026

Overview of 21cm Experiments at high redshift with SKAO

We provide an overview of the eight SKAO Science Book chapters that motivate the Epoch of Reionisation and Cosmic Dawn experiments with SKA-Low. We describe the individual SKA-Low experiments and expected sensitivity - power spectrum, tomography, 21-cm forest, cross-correlations, building on the broad observational plan laid out in the 2015 SKA Science Book. Finally, we outline features of the telescope that will be critical for the success of EoR/CD science, e.g., beam apodization, substations, and multi-beaming.

astro-ph.CO↗

arXiv2026

Monocular Normal Estimation via Shading Sequence Estimation

Monocular normal estimation aims to estimate the normal map from a single RGB image of an object under arbitrary lights. Existing methods rely on deep models to directly predict normal maps. However, they often suffer from 3D misalignment: while the estimated normal maps may appear to have a correct appearance, the reconstructed surfaces often fail to align with the geometric details. We argue that this misalignment stems from the current paradigm: the model struggles to distinguish and reconstruct varying geometry represented in normal maps, as the differences in underlying geometry are reflected only through relatively subtle color variations. To address this issue, we propose a new paradigm that reformulates normal estimation as shading sequence estimation, where shading sequences are more sensitive to various geometric information. Building on this paradigm, we present RoSE, a method that leverages image-to-video generative models to predict shading sequences. The predicted shading sequences are then converted into normal maps by solving a simple ordinary least-squares problem. To enhance robustness and better handle complex objects, RoSE is trained on a synthetic dataset, MultiShade, with diverse shapes, materials, and light conditions. Experiments demonstrate that RoSE achieves state-of-the-art performance on real-world benchmark datasets for object-based monocular normal estimation.

cs.CV↗

arXiv2026

FedAFD: Multimodal Federated Learning via Adversarial Fusion and Distillation

Multimodal Federated Learning (MFL) enables clients with heterogeneous data modalities to collaboratively train models without sharing raw data, offering a privacy-preserving framework that leverages complementary cross-modal information. However, existing methods often overlook personalized client performance and struggle with modality/task discrepancies, as well as model heterogeneity. To address these challenges, we propose FedAFD, a unified MFL framework that enhances client and server learning. On the client side, we introduce a bi-level adversarial alignment strategy to align local and global representations within and across modalities, mitigating modality and task gaps. We further design a granularity-aware fusion module to integrate global knowledge into the personalized features adaptively. On the server side, to handle model heterogeneity, we propose a similarity-guided ensemble distillation mechanism that aggregates client representations on shared public data based on feature similarity and distills the fused knowledge into the global model. Extensive experiments conducted under both IID and non-IID settings demonstrate that FedAFD achieves superior performance and efficiency for both the client and the server.

cs.LG↗

arXiv2026

Simulation and Data Processing of Beamforming Experiments with Four 21CMA Stations

We present an end-to-end simulation and data-processing framework for digital beamforming experiments conducted with four stations of the 21 Centimeter Array (21CMA). Motivated by the need to characterize instrumental systematics, such as those arising from station-level digital beam synthesis and two-stage channelization, and to validate the data-processing pipeline framework for a future upgraded 21CMA with beamforming capability across all stations, we simulate interferometric visibilities using realistic four-station layouts with radio interferometer simulation software OSKAR. Two representative pointings are considered: a bright, complex Cassiopeia A field and a near-north celestial pole (NCP) calibration field. The sky model combines cataloged point sources with a diffuse Galactic component from the Global Sky Model (GSM), and frequency-dependent thermal noise is injected. We further quantify the imprint of two-stage channelization by comparing an ideal beamformer with a coarse-channel phase approximation, demonstrating that off-axis sources exhibit a characteristic piecewise-linear spectral modulation across coarse-channel boundaries. A data-processing pipeline, including Radio Frequency Interference (RFI) mitigation, calibration, imaging, and mosaicking steps consistent with current low-frequency radio astronomy practice, is constructed. The resulting synthetic images and background root-mean-square (RMS) noise measurements demonstrate the feasibility of adapting established 21CMA calibration and imaging strategies to digital beamforming modes, and provide a framework that can be further developed for beam-aware processing in future full-scale 21CMA beamforming observations.

astro-ph.IM↗

arXiv2026

Neural-Driven Image Editing

Traditional image editing typically relies on manual prompting, making it labor-intensive and inaccessible to individuals with limited motor control or language abilities. Leveraging recent advances in brain-computer interfaces (BCIs) and generative models, we propose LoongX, a hands-free image editing approach driven by multimodal neurophysiological signals. LoongX utilizes state-of-the-art diffusion models trained on a comprehensive dataset of 23,928 image editing pairs, each paired with synchronized electroencephalography (EEG), functional near-infrared spectroscopy (fNIRS), photoplethysmography (PPG), and head motion signals that capture user intent. To effectively address the heterogeneity of these signals, LoongX integrates two key modules. The cross-scale state space (CS3) module encodes informative modality-specific features. The dynamic gated fusion (DGF) module further aggregates these features into a unified latent space, which is then aligned with edit semantics via fine-tuning on a diffusion transformer (DiT). Additionally, we pre-train the encoders using contrastive learning to align cognitive states with semantic intentions from embedded natural language. Extensive experiments demonstrate that LoongX achieves performance comparable to text-driven methods (CLIP-I: 0.6605 vs. 0.6558; DINO: 0.4812 vs. 0.4636) and outperforms them when neural signals are combined with speech (CLIP-T: 0.2588 vs. 0.2549). These results highlight the promise of neural-driven generative models in enabling accessible, intuitive image editing and open new directions for cognitive-driven creative technologies. The code and dataset are released on the project website: https://loongx1.github.io.

cs.CV↗

arXiv2025

DD-Ranking: Rethinking the Evaluation of Dataset Distillation

In recent years, dataset distillation has provided a reliable solution for data compression, where models trained on the resulting smaller synthetic datasets achieve performance comparable to those trained on the original datasets. To further improve the performance of synthetic datasets, various training pipelines and optimization objectives have been proposed, greatly advancing the field of dataset distillation. Recent decoupled dataset distillation methods introduce soft labels and stronger data augmentation during the post-evaluation phase and scale dataset distillation up to larger datasets (e.g., ImageNet-1K). However, this raises a question: Is accuracy still a reliable metric to fairly evaluate dataset distillation methods? Our empirical findings suggest that the performance improvements of these methods often stem from additional techniques rather than the inherent quality of the images themselves, with even randomly sampled images achieving superior results. Such misaligned evaluation settings severely hinder the development of DD. Therefore, we propose DD-Ranking, a unified evaluation framework, along with new general evaluation metrics to uncover the true performance improvements achieved by different methods. By refocusing on the actual information enhancement of distilled datasets, DD-Ranking provides a more comprehensive and fair evaluation standard for future research advancements.

cs.CV↗

arXiv2025

DarwinWafer: A Wafer-Scale Neuromorphic Chip

Neuromorphic computing promises brain-like efficiency, yet today's multi-chip systems scale over PCBs and incur orders-of-magnitude penalties in bandwidth, latency, and energy, undermining biological algorithms and system efficiency. We present DarwinWafer, a hyperscale system-on-wafer that replaces off-chip interconnects with wafer-scale, high-density integration of 64 Darwin3 chiplets on a 300 mm silicon interposer. A GALS NoC within each chiplet and an AER-based asynchronous wafer fabric with hierarchical time-step synchronization provide low-latency, coherent operation across the wafer. Each chiplet implements 2.35 M neurons and 0.1 B synapses, yielding 0.15 B neurons and 6.4 B synapses per wafer.At 333 MHz and 0.8 V, DarwinWafer consumes ~100 W and achieves 4.9 pJ/SOP, with 64 TSOPS peak throughput (0.64 TSOPS/W). Realization is enabled by a holistic chiplet-interposer co-design flow (including an in-house interposer-bump planner with early SI/PI and electro-thermal closure) and a warpage-tolerant assembly that fans out I/O via PCBlets and compliant pogo-pin connections, enabling robust, demountable wafer-to-board integration. Measurements confirm 10 mV supply droop and a uniform thermal profile (34-36 °C) under ~100 W. Application studies demonstrate whole-brain simulations: two zebrafish brains per chiplet with high connectivity fidelity (Spearman r = 0.896) and a mouse brain mapped across 32 chiplets (r = 0.645). To our knowledge, DarwinWafer represents a pioneering demonstration of wafer-scale neuromorphic computing, establishing a viable and scalable path toward large-scale, brain-like computation on silicon by replacing PCB-level interconnects with high-density, on-wafer integration.

cs.ET↗

arXiv2025

A candidate field for deep imaging of the Epoch of Reionization observed with MWA

Deep imaging of structures from the Cosmic Dawn (CD) and the Epoch of Reionization (EoR) in five targeted fields is one of the highest priority scientific objectives for the Square Kilometre Array (SKA). Selecting 'quiet' fields, which allow deep imaging, is critical for future SKA CD/EoR observations. Pre-observations using existing radio facilities will help estimate the computational capabilities required for optimal data quality and refine data reduction techniques. In this study, we utilize data from the Murchison Widefield Array (MWA) Phase II extended array for a selected field to study the properties of foregrounds. We conduct deep imaging across two frequency bands: 72-103 MHz and 200-231 MHz. We identify up to 2,576 radio sources within a 5-degree radius of the image center (at RA (J2000) $8^h$ , Dec (J2000) 5°), achieving approximately 80% completeness at 7.7 mJy and 90% at 10.4 mJy for 216 MHz, with a total integration time of 4.43 hours and an average RMS of 1.80 mJy. Additionally, we apply a foreground removal algorithm using Principal Component Analysis (PCA) and calculate the angular power spectra of the residual images. Our results indicate that nearly all resolved radio sources can be successfully removed using PCA, leading to a reduction in foreground power. However, the angular power spectra of the residual map remains over an order of magnitude higher than the theoretically predicted CD/EoR 21 cm signal. Further improvements in data reduction and foreground subtraction techniques will be necessary to enhance these results.

astro-ph.IM↗

arXiv2025

MWA and VLA Observations of Diffuse Radio Lobes in M 87

This study investigates the projected, quasi-symmetric $\sim\rm46\,kpc$-scale diffuse radio lobes surrounding the giant elliptical galaxy M\,87, utilizing well-sampled wideband ($\rm 60\,MHz-10.55\,GHz$) observations from MWA and VLA, supplemented by data from LOFAR and Effelsberg. The observed structures feature sharp edges and filaments, with nearly uniform and moderately steep spectral indices ($α$, mostly within $-1.2\leqα\leq-0.8$), indicating turbulence. Well-sampled radio spectra for the lobes' diffuse region are derived using the continuous injection (CI) model (with $α_{\rm inj}\simeq-0.86$ and $ν_{\rm b}\simeq1.72\rm\,GHz$), and for its three localized regions using the impulsive injection model (e.g., JP model). From energy equipartition analysis, we estimate the typical magnetic field strength in the lobes' diffuse region to be $B_{\rm eq}\simeq10\,μ\rm G$. The age of the lobes is estimated as $\sim30-50\,\rm~Myr$, based on lifetimes derived from the CI and JP models and sound crossing time. Outflow powers of $\sim(0.2-2)\times10^{44}\,\rm erg\,s^{-1}$ for the lobes' diffuse components and $\sim(1-11)\times10^{44}\,\rm erg\,s^{-1}$ for the whole source are calculated. With this power assessment, we conclude that the galactic stellar wind has a negligible effect, the active galactic nucleus (AGN)-driven jet can provide the necessary energy for the whole system. Furthermore, we argue that while the wind driven by current AGN activity is unlikely to power the lobes' diffuse components, an average enhancement of AGN activity by a factor of $\sim 10^2$ over the past $\sim 30-50$ Myr remains plausible.

astro-ph.GA↗

arXiv2025

Low-bit Model Quantization for Deep Neural Networks: A Survey

With unprecedented rapid development, deep neural networks (DNNs) have deeply influenced almost all fields. However, their heavy computation costs and model sizes are usually unacceptable in real-world deployment. Model quantization, an effective weight-lighting technique, has become an indispensable procedure in the whole deployment pipeline. The essence of quantization acceleration is the conversion from continuous floating-point numbers to discrete integer ones, which significantly speeds up the memory I/O and calculation, i.e., addition and multiplication. However, performance degradation also comes with the conversion because of the loss of precision. Therefore, it has become increasingly popular and critical to investigate how to perform the conversion and how to compensate for the information loss. This article surveys the recent five-year progress towards low-bit quantization on DNNs. We discuss and compare the state-of-the-art quantization methods and classify them into 8 main categories and 24 sub-categories according to their core techniques. Furthermore, we shed light on the potential research opportunities in the field of model quantization. A curated list of model quantization is provided at https://github.com/Kai-Liu001/Awesome-Model-Quantization.

cs.LG↗

arXiv2025

The Short-spacing Interferometer Array for Global 21-cm Signal Detection (SIGMA): Design of the Antennas and Layout

Numerous experiments have been designed to investigate the Cosmic Dawn (CD) and Epoch of Reionization (EoR) by examining redshifted 21-cm emissions from neutral hydrogen. Detecting the global spectrum of redshifted 21-cm signals is typically achieved through single-antenna experiments. However, this global 21-cm signal is deeply embedded in foreground emissions, which are about four orders of magnitude stronger. Extracting this faint signal is a significant challenge, requiring highly precise instrumental calibration. Additionally, accurately modelling receiver noise in single-antenna experiments is inherently complex. An alternative approach using a short-spacing interferometer is expected to alleviate these difficulties because the noise in different receivers is uncorrelated and averages to zero upon cross-correlation. The Short-spacing Interferometer array for Global 21-cm Signal detection (SIGMA) is an upcoming experiment aimed at detecting the global CD/EoR signal using this approach. We describe the SIGMA system with a focus on optimal antenna design and layout, and propose a framework to address cross-talk between antennas in future calibrations. The SIGMA system is intended to serve as a prototype to gain a better understanding of the system's instrumental effects and to optimize its performance further.

astro-ph.IM↗

arXiv2025

Mitigating Reward Over-Optimization in RLHF via Behavior-Supported Regularization

Reinforcement learning from human feedback (RLHF) is an effective method for aligning large language models (LLMs) with human values. However, reward over-optimization remains an open challenge leading to discrepancies between the performance of LLMs under the reward model and the true human objectives. A primary contributor to reward over-optimization is the extrapolation error that arises when the reward model evaluates out-of-distribution (OOD) responses. However, current methods still fail to prevent the increasing frequency of OOD response generation during the reinforcement learning (RL) process and are not effective at handling extrapolation errors from OOD responses. In this work, we propose the Behavior-Supported Policy Optimization (BSPO) method to mitigate the reward over-optimization issue. Specifically, we define behavior policy as the next token distribution of the reward training dataset to model the in-distribution (ID) region of the reward model. Building on this, we introduce the behavior-supported Bellman operator to regularize the value function, penalizing all OOD values without impacting the ID ones. Consequently, BSPO reduces the generation of OOD responses during the RL process, thereby avoiding overestimation caused by the reward model's extrapolation errors. Theoretically, we prove that BSPO guarantees a monotonic improvement of the supported policy until convergence to the optimal behavior-supported policy. Empirical results from extensive experiments show that BSPO outperforms baselines in preventing reward over-optimization due to OOD evaluation and finding the optimal ID policy.

cs.LG↗

arXiv2025

MeerKAT discovery of GHz radio emission extending from Abell 3017 toward Abell 3016

Context: The clusters Abell 3017 and Abell 3016 are located within a large-scale filament. A prominent X-ray bridge has been detected connecting the two clusters and a potential galaxy group between them. Aims: The aim of this work is to investigate the existence of a radio bridge in the filament between Abell 3017 and Abell 3016, to explore other diffuse radio structures within this system, and to investigate the origins of these diffuse radio emission. Methods: We analyzed MeerKAT L-band data to study the morphology and spectra of the diffuse radio structures in Abell 3016-Abell 3017. X-ray imaging and spectral analysis were carried out with archival Chandra and XMM-Newton data. Additionally, correlations between radio ($I_R$) and X-ray surface brightness ($I_X$) were generated to explore the connections between thermal and non-thermal components in the diffuse radio emission. Results: We detected a faint radio bridge with an average surface brightness of $\sim 0.1~μ\rm Jy~arcsec^{-2}$ at 1280 MHz using MeerKAT. It connects Abell 3017 with a potential galaxy group and extends towards Abell 3016, aligning with the X-ray bridge. A high X-ray temperature of $7.09 \pm 0.54$ keV detected in the bridge region suggests an interaction between Abell 3017 and the group. In Abell 3017, we identified two distinct components of diffuse radio emission: a radio mini-halo and an outer radio halo with a northern extension (N-extension hereafter). The radio surface brightness profile of Abell 3017 shows a steep inner component consistent with other mini-halos, and a faint outer component likely linked to an infalling subcluster. The $I_{\rm R}-I_{\rm X}$ diagram indicates superlinear and sublinear correlations for the mini-halo and N-extension, respectively.

astro-ph.CO↗

arXiv2024

Darkit: A User-Friendly Software Toolkit for Spiking Large Language Model

Large language models (LLMs) have been widely applied in various practical applications, typically comprising billions of parameters, with inference processes requiring substantial energy and computational resources. In contrast, the human brain, employing bio-plausible spiking mechanisms, can accomplish the same tasks while significantly reducing energy consumption, even with a similar number of parameters. Based on this, several pioneering researchers have proposed and implemented various large language models that leverage spiking neural networks. They have demonstrated the feasibility of these models, validated their performance, and open-sourced their frameworks and partial source code. To accelerate the adoption of brain-inspired large language models and facilitate secondary development for researchers, we are releasing a software toolkit named DarwinKit (Darkit). The toolkit is designed specifically for learners, researchers, and developers working on spiking large models, offering a suite of highly user-friendly features that greatly simplify the learning, deployment, and development processes.

cs.SE↗

arXiv2024

Safe Reinforcement Learning using Finite-Horizon Gradient-based Estimation

A key aspect of Safe Reinforcement Learning (Safe RL) involves estimating the constraint condition for the next policy, which is crucial for guiding the optimization of safe policy updates. However, the existing Advantage-based Estimation (ABE) method relies on the infinite-horizon discounted advantage function. This dependence leads to catastrophic errors in finite-horizon scenarios with non-discounted constraints, resulting in safety-violation updates. In response, we propose the first estimation method for finite-horizon non-discounted constraints in deep Safe RL, termed Gradient-based Estimation (GBE), which relies on the analytic gradient derived along trajectories. Our theoretical and empirical analyses demonstrate that GBE can effectively estimate constraint changes over a finite horizon. Constructing a surrogate optimization problem with GBE, we developed a novel Safe RL algorithm called Constrained Gradient-based Policy Optimization (CGPO). CGPO identifies feasible optimal policies by iteratively resolving sub-problems within trust regions. Our empirical results reveal that CGPO, unlike baseline algorithms, successfully estimates the constraint functions of subsequent policies, thereby ensuring the efficiency and feasibility of each update.

cs.LG↗

arXiv2024

Spiking GS: Towards High-Accuracy and Low-Cost Surface Reconstruction via Spiking Neuron-based Gaussian Splatting

3D Gaussian Splatting is capable of reconstructing 3D scenes in minutes. Despite recent advances in improving surface reconstruction accuracy, the reconstructed results still exhibit bias and suffer from inefficiency in storage and training. This paper provides a different observation on the cause of the inefficiency and the reconstruction bias, which is attributed to the integration of the low-opacity parts (LOPs) of the generated Gaussians. We show that LOPs consist of Gaussians with overall low-opacity (LOGs) and the low-opacity tails (LOTs) of Gaussians. We propose Spiking GS to reduce such two types of LOPs by integrating spiking neurons into the Gaussian Splatting pipeline. Specifically, we introduce global and local full-precision integrate-and-fire spiking neurons to the opacity and representation function of flattened 3D Gaussians, respectively. Furthermore, we enhance the density control strategy with spiking neurons' thresholds and a new criterion on the scale of Gaussians. Our method can represent more accurate reconstructed surfaces at a lower cost. The supplementary material and code are available at https://github.com/zju-bmi-lab/SpikingGS.

cs.CV↗