arXiv subjects

Tao Wang

Publications and source records attributed to Tao Wang.

At least 271 records · Page 15Linked to original sources

arXiv2025

Rydberg atomic spectrum analyzer with microwave-dressed-state-locking and multimode Floquet theory

We propose a Rydberg atomic spectrum analyzer (RASA) utilizing microwave-dressed-state-locking (MWDSL) in conjunction with multimode Floquet theory (MFT). By leveraging a strong local microwave (MW) field resonant with Rydberg states to implement MWDSL, we analyze the second-order effect of MFT induced by the interplay of controllable bias and signal MW fields. This effect facilitates the coupling of locked dressed states, providing a pathway for measuring the signal MW field. We demonstrate that the RASA can simultaneously characterize multiple MW fields across distinct frequencies, with both the frequency and strength of each MW field discernible in the spectral response. This capability renders RASA suitable for measuring unknown-frequency MW fields, thereby expanding the utility of Rydberg atom-based electrometers in complex spectral analysis scenarios.

physics.atom-ph↗

arXiv2025

Collaborative Multi-LoRA Experts with Achievement-based Multi-Tasks Loss for Unified Multimodal Information Extraction

Multimodal Information Extraction (MIE) has gained attention for extracting structured information from multimedia sources. Traditional methods tackle MIE tasks separately, missing opportunities to share knowledge across tasks. Recent approaches unify these tasks into a generation problem using instruction-based T5 models with visual adaptors, optimized through full-parameter fine-tuning. However, this method is computationally intensive, and multi-task fine-tuning often faces gradient conflicts, limiting performance. To address these challenges, we propose collaborative multi-LoRA experts with achievement-based multi-task loss (C-LoRAE) for MIE tasks. C-LoRAE extends the low-rank adaptation (LoRA) method by incorporating a universal expert to learn shared multimodal knowledge from cross-MIE tasks and task-specific experts to learn specialized instructional task features. This configuration enhances the model's generalization ability across multiple tasks while maintaining the independence of various instruction tasks and mitigating gradient conflicts. Additionally, we propose an achievement-based multi-task loss to balance training progress across tasks, addressing the imbalance caused by varying numbers of training samples in MIE tasks. Experimental results on seven benchmark datasets across three key MIE tasks demonstrate that C-LoRAE achieves superior overall performance compared to traditional fine-tuning methods and LoRA methods while utilizing a comparable number of training parameters to LoRA.

cs.LG↗

arXiv2025

Identification of BHB stars using Synthetic SkyMapper colors from Gaia XP spectra

Blue horizontal-branch (BHB) stars are ideal tracers for mapping the structure of Galactic stellar halo. Traditionally, BHB sample stars are built from large-scale spectroscopic surveys utilizing their spectral features, however, the resulting sample sizes have been quite limited. In this paper, we construct a catalog of BHB stars based on synthetic colors $(u-v)_{0}$ and $(g-i)_{0}$ in SkyMapper photometric systems, which are convolved from Gaia XP spectra. A total of 49,733 BHB stars are selected from nearly the entire sky (excluding regions of low Galactic latitudes $|b| \le 8^{\circ}$ with heavy reddening), with a completeness and purity exceeding 90\%. Using member stars of globular clusters with precise distance determinations, we carefully calibrate the relationship between the $g$-band absolute magnitude and $(g-i)_{0}$, achieving a precision of 0.11\,mag, which corresponds to a 5\% uncertainty in distance. This relation is applied to derive distances for all BHB stars in the constructed sample. Given current capabilities of Gaia XP observations, the constructed BHB sample is primarily located within 20 kpc, enabling detailed mapping of the inner stellar halo. To extend this depth to the outer halo or even the edge of our Galaxy, we explore the potential of the Chinese Space Station Telescope (CSST) and its broad-band photometry for detecting BHB stars. Using mock data from synthetic spectra, we find that it is feasible to distinguish BHB stars from blue stragglers (BS) stars using CSST near-ultraviolet bands ($NUV, u$) photometry. Thanks to the deep limiting magnitude of CSST, its data will provide a groundbreaking perspective on our Galaxy, particularly regarding the outer halo, in an unprecedented volume.

astro-ph.GA↗

arXiv2025

P2Mark: Plug-and-play Parameter-level Watermarking for Neural Speech Generation

Neural speech generation (NSG) has rapidly advanced as a key component of artificial intelligence-generated content, enabling the generation of high-quality, highly realistic speech for diverse applications. This development increases the risk of technique misuse and threatens social security. Audio watermarking can embed imperceptible marks into generated audio, providing a promising approach for secure NSG usage. However, current audio watermarking methods are mainly applied at the audio-level or feature-level, which are not suitable for open-sourced scenarios where source codes and model weights are released. To address this limitation, we propose a Plug-and-play Parameter-level WaterMarking (P2Mark) method for NSG. Specifically, we embed watermarks into the released model weights, offering a reliable solution for proactively tracing and protecting model copyrights in open-source scenarios. During training, we introduce a lightweight watermark adapter into the pre-trained model, allowing watermark information to be merged into the model via this adapter. This design ensures both the flexibility to modify the watermark before model release and the security of embedding the watermark within model parameters after model release. Meanwhile, we propose a gradient orthogonal projection optimization strategy to ensure the quality of the generated audio and the accuracy of watermark preservation. Experimental results on two mainstream waveform decoders in NSG (i.e., vocoder and codec) demonstrate that P2Mark achieves comparable performance to state-of-the-art audio watermarking methods that are not applicable to open-source white-box protection scenarios, in terms of watermark extraction accuracy, watermark imperceptibility, and robustness.

cs.SD↗

arXiv2025

Precision Neural Network Quantization via Learnable Adaptive Modules

Quantization Aware Training (QAT) is a neural network quantization technique that compresses model size and improves operational efficiency while effectively maintaining model performance. The paradigm of QAT is to introduce fake quantization operators during the training process, allowing the model to autonomously compensate for information loss caused by quantization. Making quantization parameters trainable can significantly improve the performance of QAT, but at the cost of compromising the flexibility during inference, especially when dealing with activation values with substantially different distributions. In this paper, we propose an effective learnable adaptive neural network quantization method, called Adaptive Step Size Quantization (ASQ), to resolve this conflict. Specifically, the proposed ASQ method first dynamically adjusts quantization scaling factors through a trained module capable of accommodating different activations. Then, to address the rigid resolution issue inherent in Power of Two (POT) quantization, we propose an efficient non-uniform quantization scheme. We utilize the Power Of Square root of Two (POST) as the basis for exponential quantization, effectively handling the bell-shaped distribution of neural network weights across various bit-widths while maintaining computational efficiency through a Look-Up Table method (LUT). Extensive experimental results demonstrate that the proposed ASQ method is superior to the state-of-the-art QAT approaches. Notably that the ASQ is even competitive compared to full precision baselines, with its 4-bit quantized ResNet34 model improving accuracy by 1.2\% on ImageNet.

cs.CV↗

arXiv2025

Memory-Augmented Dual-Decoder Networks for Multi-Class Unsupervised Anomaly Detection

Recent advances in unsupervised anomaly detection (UAD) have shifted from single-class to multi-class scenarios. In such complex contexts, the increasing pattern diversity has brought two challenges to reconstruction-based approaches: (1) over-generalization: anomalies that are subtle or share compositional similarities with normal patterns may be reconstructed with high fidelity, making them difficult to distinguish from normal instances; and (2) insufficient normality reconstruction: complex normal features, such as intricate textures or fine-grained structures, may not be faithfully reconstructed due to the model's limited representational capacity, resulting in false positives. Existing methods typically focus on addressing the former, which unintentionally exacerbate the latter, resulting in inadequate representation of intricate normal patterns. To concurrently address these two challenges, we propose a Memory-augmented Dual-Decoder Networks (MDD-Net). This network includes two critical components: a Dual-Decoder Reverse Distillation Network (DRD-Net) and a Class-aware Memory Module (CMM). Specifically, the DRD-Net incorporates a restoration decoder designed to recover normal features from synthetic abnormal inputs and an identity decoder to reconstruct features that maintain the anomalous semantics. By exploiting the discrepancy between features produced by two decoders, our approach refines anomaly scores beyond the conventional encoder-decoder comparison paradigm, effectively reducing false positives and enhancing localization accuracy. Furthermore, the CMM explicitly encodes and preserves class-specific normal prototypes, actively steering the network away from anomaly reconstruction. Comprehensive experimental results across several benchmarks demonstrate the superior performance of our MDD-Net framework over current SoTA approaches in multi-class UAD tasks.

cs.CV↗

arXiv2025

ToReMi: Topic-Aware Data Reweighting for Dynamic Pre-Training Data Selection

Pre-training large language models (LLMs) necessitates enormous diverse textual corpora, making effective data selection a key challenge for balancing computational resources and model performance. Current methodologies primarily emphasize data quality metrics and mixing proportions, yet they fail to adequately capture the underlying semantic connections between training samples and quality disparities within individual domains. We introduce ToReMi (Topic-based Reweighting for Model improvement), a novel two-stage framework that dynamically adjusts training sample weights according to their topical associations and observed learning patterns. Our comprehensive experiments reveal that ToReMi variants consistently achieve superior performance over conventional pre-training approaches, demonstrating accelerated perplexity reduction across multiple domains and enhanced capabilities on downstream evaluation tasks. Code is available at https://github.com/zxx000728/ToReMi.

cs.CL↗

arXiv2025

Level-anticrossing in B(E2) anomaly (I)

Recently, a new mechanism for explaining the B(E2) anomaly was given by F. Pan \emph{et al.} (PRC, 110, 054324, 2024), which is realized in the parameter region from the SU(3) symmetry limit to the O(6) symmetry limit, and seems to be not related to the SU(3) symmetry. However, through SU(3) analysis, a new technique proposed recently, we found that it is not so. The new mechanism is related to level-anticrossing phenomenon, which is related to level-crossing phenomenon in the SU(3) symmetry limit. By incorporating previous ideas, we have a more general explanatory framework for the B(E2) anomaly, which is important for understanding some higher-order interactions in the interacting boson model. Through analysis, it is shown that level-anticrossing in this mechanism mainly results from the third-order interaction $[\hat{L}\times \hat{Q}_χ \times \hat{L}]^{(0)}$. Finally, the B(E2) anomaly in $^{170}$Os is also discussed within this general framework.

nucl-th↗

arXiv2025

NTIRE 2025 Challenge on Cross-Domain Few-Shot Object Detection: Methods and Results

Cross-Domain Few-Shot Object Detection (CD-FSOD) poses significant challenges to existing object detection and few-shot detection models when applied across domains. In conjunction with NTIRE 2025, we organized the 1st CD-FSOD Challenge, aiming to advance the performance of current object detectors on entirely novel target domains with only limited labeled data. The challenge attracted 152 registered participants, received submissions from 42 teams, and concluded with 13 teams making valid final submissions. Participants approached the task from diverse perspectives, proposing novel models that achieved new state-of-the-art (SOTA) results under both open-source and closed-source settings. In this report, we present an overview of the 1st NTIRE 2025 CD-FSOD Challenge, highlighting the proposed solutions and summarizing the results submitted by the participants.

cs.CV↗

arXiv2025

Double shape quantum phase transitions in the SU3-IBM: new $γ$-soft phase and the shape phase transition from the new $γ$-soft phase to the prolate shape

Shape quantum phase transition is an important topic in nuclear structure. In this paper, we begin to study the shape quantum phase transition in the SU3-IBM. In this new proposed model, spherical-like spectra was found to resolve the spherical nucleus puzzle, which is a new $γ$-soft rotational mode. In this paper, the shape phase transition along the new $γ$-soft line is first discussed, and then the neighbouring case at the prolate side is also studied. We find that double shape phase transitions occur along a single parameter path. The new $γ$-softness is really a shape phase and the shape phase transition from the new $γ$-soft phase to the prolate shape is found. The experimental support is also found and $^{108}$Pd may be the critical nucleus.

nucl-th↗

arXiv2025

Voltage and power-frequency electric field measurements with Rydberg-atom interferometry

We present a Rydberg-atom interferometry-based technique for voltage measurement between electrodes embedded in an atomic vapor cell, enabling the detection of weak voltages ($<0.1$V) and unambiguous discrimination between positive and negative polarities. This makes up for the shortcomings of measurements based on the Stark effect, which suffer from quadratic field dependence (limiting sensitivity in weak-field regimes) and incapable of distinguishing the electric field direction. Furthermore, this method extends naturally to power-frequency (PF) electric field measurements by exploiting the quasi-static approximation-valid given the PF field's characteristic timescale ($\sim10^{-2}$s) vastly exceeds the interferometric measurement duration ($\sim10^{-6}$s). Crucially, our protocol provides instantaneous PF field reconstruction, providing comprehensive information including amplitude, frequency and phase. These advancements have direct implications for traceable voltage measurements and non-invasive characterization of PF fields near high-voltage infrastructure.

physics.atom-ph↗

arXiv2025

SU(3) analysis for B(E2) anomaly

B(E2) anomaly is becoming a hot topic in the field of nuclear structure. Since the B(E2) anomaly was experimentally found, understanding the mechanism of its production has become an important problem. Theoretical studies have found that the SU3-IBM and other extended IBM theories can give explanations, but different mechanisms were found. In this paper, the concept ``SU(3) analysis'' for B(E2) anomaly is proposed, and some new conclusions are obtained. The SU(3) third-order interaction $[L\times Q \times L]^{(0)}$ and level-crossing are both vital for the emergence of the B(E2) anomaly. This technique can help us to better understand the realistic reason of the B(E2) anomaly.

nucl-th↗

arXiv2025

Selective Masking Adversarial Attack on Automatic Speech Recognition Systems

Extensive research has shown that Automatic Speech Recognition (ASR) systems are vulnerable to audio adversarial attacks. Current attacks mainly focus on single-source scenarios, ignoring dual-source scenarios where two people are speaking simultaneously. To bridge the gap, we propose a Selective Masking Adversarial attack, namely SMA attack, which ensures that one audio source is selected for recognition while the other audio source is muted in dual-source scenarios. To better adapt to the dual-source scenario, our SMA attack constructs the normal dual-source audio from the muted audio and selected audio. SMA attack initializes the adversarial perturbation with a small Gaussian noise and iteratively optimizes it using a selective masking optimization algorithm. Extensive experiments demonstrate that the SMA attack can generate effective and imperceptible audio adversarial examples in the dual-source scenario, achieving an average success rate of attack of 100% and signal-to-noise ratio of 37.15dB on Conformer-CTC, outperforming the baselines.

cs.CR↗

arXiv2025

Stellar parameters, extinction, and distances for stars in SMSS DR2 by SPar method

The availability of large datasets containing stellar parameters, distances, and extinctions for stars in the Milky Way, particularly within the Galactic disk, is essential for advancing our understanding of the Galaxy's stellar populations, structure, kinematics, and chemical evolution. In this study, we present a catalog of stellar parameters, including effective temperature (\teff), metallicity (\feh), absolute magnitudes ($M_{G}$), distances ($d$), and reddening values (\ebr), for a sample of 141 million stars from the SkyMapper Southern Survey (SMSS). These parameters are derived using the SPar algorithm, which employs a fitting procedure to match multi-band photometric observations and estimate the stellar properties (\teff, \feh, $M_G$, $d$, and \ebr) on an individual star basis, following the methodology outlined in our previous work. This study successfully determines stellar parameters and extinction values simultaneously for stars located in high and low Galactic latitudes. The resulting stellar parameters and extinction values are in good agreement with results from other studies, demonstrating the robustness of our method. We derive a temperature dispersion of 195\,K and a metallicity dispersion of 0.31\,dex when comparing our results with spectroscopic data. The catalog produced in this work provides a valuable resource for future research on the Galactic metallicity distribution function, the structure of the Galaxy, three-dimensional extinction mapping, and other related astrophysical topics.

astro-ph.GA↗

arXiv2025

LITE: LLM-Impelled efficient Taxonomy Evaluation

This paper presents LITE, an LLM-based evaluation method designed for efficient and flexible assessment of taxonomy quality. To address challenges in large-scale taxonomy evaluation, such as efficiency, fairness, and consistency, LITE adopts a top-down hierarchical evaluation strategy, breaking down the taxonomy into manageable substructures and ensuring result reliability through cross-validation and standardized input formats. LITE also introduces a penalty mechanism to handle extreme cases and provides both quantitative performance analysis and qualitative insights by integrating evaluation metrics closely aligned with task objectives. Experimental results show that LITE demonstrates high reliability in complex evaluation tasks, effectively identifying semantic errors, logical contradictions, and structural flaws in taxonomies, while offering directions for improvement. Code is available at https://github.com/Zhang-l-i-n/TAXONOMY_DETECT .

cs.CL↗

arXiv2025

GAPO: Learning Preferential Prompt through Generative Adversarial Policy Optimization

Recent advances in large language models have highlighted the critical need for precise control over model outputs through predefined constraints. While existing methods attempt to achieve this through either direct instruction-response synthesis or preferential response optimization, they often struggle with constraint understanding and adaptation. This limitation becomes particularly evident when handling fine-grained constraints, leading to either hallucination or brittle performance. We introduce Generative Adversarial Policy Optimization (GAPO), a novel framework that combines GAN-based training dynamics with an encoder-only reward model to progressively learn and adapt to increasingly complex constraints. GAPO leverages adversarial training to automatically generate training samples of varying difficulty while utilizing the encoder-only architecture to better capture prompt-response relationships. Extensive experiments demonstrate GAPO's superior performance across multiple benchmarks, particularly in scenarios requiring fine-grained constraint handling, where it significantly outperforms existing methods like PPO, DPO, and KTO. Our results suggest that GAPO's unique approach to preferential prompt learning offers a more robust and effective solution for controlling LLM outputs. Code is avaliable in https://github.com/MikeGu721/GAPO.

cs.CL↗

arXiv2025

High-rate self-referenced continuous-variable quantum key distribution over high-loss free-space channel

The advent of quantum computers has significantly challenged the security of traditional cryptographic systems, prompting a surge in research on quantum key distribution (QKD). Among various QKD approaches, continuous-variable QKD (CVQKD) offers superior resilience against background noise. However, the local local oscillator (LLO) CVQKD scheme faces substantial physical limitations in scenarios with high channel attenuation, and the large attenuation CVQKD remains unrealized. Bottleneck challenges include ensuring stable low-noise transmission and accurately estimating parameters under fluctuating channel conditions. In this paper, we introduce a continuous-time mode theory for high-precision estimation of time-varying parameters and design a free-space experimental system with a main quantum system and an auxiliary counterpart. We further develop advanced digital signal post-processing techniques for compensating time-varying frequency offset and phase noise under dynamic channel. Notably, the estimation of the time-varying free-space channel is achieved through the use of the auxiliary quantum system. Through experimental validation, we first demonstrate high-rate secure quantum key distribution over high-loss free-space channels. Specifically, we achieve asymptotic key rates of 76.366 kbps and 403.896 kbps in 25 dB attenuation free-space channels without turbulence and 21.5 dB average attenuation free-space channels with turbulence, respectively. Additionally, we confirm the feasibility of experiments on mildly turbulent atmospheric channels spanning at least 10.5 km using current equipments. Our scheme provides direct insight into constructing an integrated air-ground quantum communication network.

quant-ph↗

arXiv2025

Segmentation Guided Sparse Transformer for Under-Display Camera Image Restoration

Under-Display Camera (UDC) is an emerging technology that achieves full-screen display via hiding the camera under the display panel. However, the current implementation of UDC causes serious degradation. The incident light required for camera imaging undergoes attenuation and diffraction when passing through the display panel, leading to various artifacts in UDC imaging. Presently, the prevailing UDC image restoration methods predominantly utilize convolutional neural network architectures, whereas Transformer-based methods have exhibited superior performance in the majority of image restoration tasks. This is attributed to the Transformer's capability to sample global features for the local reconstruction of images, thereby achieving high-quality image restoration. In this paper, we observe that when using the Vision Transformer for UDC degraded image restoration, the global attention samples a large amount of redundant information and noise. Furthermore, compared to the ordinary Transformer employing dense attention, the Transformer utilizing sparse attention can alleviate the adverse impact of redundant information and noise. Building upon this discovery, we propose a Segmentation Guided Sparse Transformer method (SGSFormer) for the task of restoring high-quality images from UDC degraded images. Specifically, we utilize sparse self-attention to filter out redundant information and noise, directing the model's attention to focus on the features more relevant to the degraded regions in need of reconstruction. Moreover, we integrate the instance segmentation map as prior information to guide the sparse self-attention in filtering and focusing on the correct regions.

eess.IV↗