arXiv subjects

Chang Li

Publications and source records attributed to Chang Li.

At least 55 records · Page 3Linked to original sources

arXiv2025

FreeAudio: Training-Free Timing Planning for Controllable Long-Form Text-to-Audio Generation

Text-to-audio (T2A) generation has achieved promising results with the recent advances in generative models. However, because of the limited quality and quantity of temporally-aligned audio-text pairs, existing T2A methods struggle to handle the complex text prompts that contain precise timing control, e.g., "owl hooted at 2.4s-5.2s". Recent works have explored data augmentation techniques or introduced timing conditions as model inputs to enable timing-conditioned 10-second T2A generation, while their synthesis quality is still limited. In this work, we propose a novel training-free timing-controlled T2A framework, FreeAudio, making the first attempt to enable timing-controlled long-form T2A generation, e.g., "owl hooted at 2.4s-5.2s and crickets chirping at 0s-24s". Specifically, we first employ an LLM to plan non-overlapping time windows and recaption each with a refined natural language description, based on the input text and timing prompts. Then we introduce: 1) Decoupling and Aggregating Attention Control for precise timing control; 2) Contextual Latent Composition for local smoothness and Reference Guidance for global consistency. Extensive experiments show that: 1) FreeAudio achieves state-of-the-art timing-conditioned T2A synthesis quality among training-free methods and is comparable to leading training-based methods; 2) FreeAudio demonstrates comparable long-form generation quality with training-based Stable Audio and paves the way for timing-controlled long-form T2A synthesis. Demo samples are available at: https://freeaudio.github.io/FreeAudio/

cs.SD↗

arXiv2025

Towards Powerful and Practical Patch Attacks for 2D Object Detection in Autonomous Driving

Learning-based autonomous driving systems remain critically vulnerable to adversarial patches, posing serious safety and security risks in their real-world deployment. Black-box attacks, notable for their high attack success rate without model knowledge, are especially concerning, with their transferability extensively studied to reduce computational costs compared to query-based attacks. Previous transferability-based black-box attacks typically adopt mean Average Precision (mAP) as the evaluation metric and design training loss accordingly. However, due to the presence of multiple detected bounding boxes and the relatively lenient Intersection over Union (IoU) thresholds, the attack effectiveness of these approaches is often overestimated, resulting in reduced success rates in practical attacking scenarios. Furthermore, patches trained on low-resolution data often fail to maintain effectiveness on high-resolution images, limiting their transferability to autonomous driving datasets. To fill this gap, we propose P$^3$A, a Powerful and Practical Patch Attack framework for 2D object detection in autonomous driving, specifically optimized for high-resolution datasets. First, we introduce a novel metric, Practical Attack Success Rate (PASR), to more accurately quantify attack effectiveness with greater relevance for pedestrian safety. Second, we present a tailored Localization-Confidence Suppression Loss (LCSL) to improve attack transferability under PASR. Finally, to maintain the transferability for high-resolution datasets, we further incorporate the Probabilistic Scale-Preserving Padding (PSPP) into the patch attack pipeline as a data preprocessing step. Extensive experiments show that P$^3$A outperforms state-of-the-art attacks on unseen models and unseen high-resolution datasets, both under the proposed practical IoU-based evaluation metric and the previous mAP-based metrics.

cs.CV↗

arXiv2025

Entanglement swapping systems toward a quantum internet

We demonstrate conditional entanglement swapping, i.e. teleportation of entanglement, between time-bin qubits at the telecommunication wavelength of 1536.4 nm with high fidelity of 87\%. Our system is deployable, utilizing modular, off-the-shelf, fiber-coupled, and electrically controlled components such as electro-optic modulators. It leverages the precise timing resolution of superconducting nanowire detectors, which are controlled and read out via a custom developed graphical user interface. The swapping process is described, interpreted, and guided using characteristic function-based analytical modeling that accounts for realistic imperfections. Our system supports quantum networking protocols, including source-independent quantum key distribution, with an estimated secret key rate of approximately 0.5 bits per sifted bit.

quant-ph↗

arXiv2025

Latent Swap Joint Diffusion for 2D Long-Form Latent Generation

This paper introduces Swap Forward (SaFa), a modality-agnostic and efficient method to generate seamless and coherence long spectrum and panorama through latent swap joint diffusion across multi-views. We first investigate the spectrum aliasing problem in spectrum-based audio generation caused by existing joint diffusion methods. Through a comparative analysis of the VAE latent representation of Mel-spectra and RGB images, we identify that the failure arises from excessive suppression of high-frequency components during the spectrum denoising process due to the averaging operator. To address this issue, we propose Self-Loop Latent Swap, a frame-level bidirectional swap applied to the overlapping region of adjacent views. Leveraging stepwise differentiated trajectories of adjacent subviews, this swap operator adaptively enhances high-frequency components and avoid spectrum distortion. Furthermore, to improve global cross-view consistency in non-overlapping regions, we introduce Reference-Guided Latent Swap, a unidirectional latent swap operator that provides a centralized reference trajectory to synchronize subview diffusions. By refining swap timing and intervals, we can achieve a cross-view similarity-diversity balance in a forward-only manner. Quantitative and qualitative experiments demonstrate that SaFa significantly outperforms existing joint diffusion methods and even training-based methods in audio generation using both U-Net and DiT models, along with effective longer length adaptation. It also adapts well to panorama generation, achieving comparable performance with 2 $\sim$ 20 $\times$ faster speed and greater model generalizability. More generation demos are available at https://swapforward.github.io/

cs.SD↗

arXiv2025

Learning Temporal Abstractions via Variational Homomorphisms in Option-Induced Abstract MDPs

Large Language Models (LLMs) have shown remarkable reasoning ability through explicit Chain-of-Thought (CoT) prompting, but generating these step-by-step textual explanations is computationally expensive and slow. To overcome this, we aim to develop a framework for efficient, implicit reasoning, where the model "thinks" in a latent space without generating explicit text for every step. We propose that these latent thoughts can be modeled as temporally-extended abstract actions, or options, within a hierarchical reinforcement learning framework. To effectively learn a diverse library of options as latent embeddings, we first introduce the Variational Markovian Option Critic (VMOC), an off-policy algorithm that uses variational inference within the HiT-MDP framework. To provide a rigorous foundation for using these options as an abstract reasoning space, we extend the theory of continuous MDP homomorphisms. This proves that learning a policy in the simplified, abstract latent space, for which VMOC is suited, preserves the optimality of the solution to the original, complex problem. Finally, we propose a cold-start procedure that leverages supervised fine-tuning (SFT) data to distill human reasoning demonstrations into this latent option space, providing a rich initialization for the model's reasoning capabilities. Extensive experiments demonstrate that our approach achieves strong performance on complex logical reasoning benchmarks and challenging locomotion tasks, validating our framework as a principled method for learning abstract skills for both language and control.

cs.AI↗

arXiv2025

A Novel Downsampling Strategy Based on Information Complementarity for Medical Image Segmentation

In convolutional neural networks (CNNs), downsampling operations are crucial to model performance. Although traditional downsampling methods (such as maximum pooling and cross-row convolution) perform well in feature aggregation, receptive field expansion, and computational reduction, they may lead to the loss of key spatial information in semantic segmentation tasks, thereby affecting the pixel-by-pixel prediction accuracy.To this end, this study proposes a downsampling method based on information complementarity - Hybrid Pooling Downsampling (HPD). The core is to replace the traditional method with MinMaxPooling, and effectively retain the light and dark contrast and detail features of the image by extracting the maximum value information of the local area.Experiment on various CNN architectures on the ACDC and Synapse datasets show that HPD outperforms traditional methods in segmentation performance, and increases the DSC coefficient by 0.5% on average. The results show that the HPD module provides an efficient solution for semantic segmentation tasks.

cs.CV↗

arXiv2025

Quality-aware Masked Diffusion Transformer for Enhanced Music Generation

Text-to-music (TTM) generation, which converts textual descriptions into audio, opens up innovative avenues for multimedia creation. Achieving high quality and diversity in this process demands extensive, high-quality data, which are often scarce in available datasets. Most open-source datasets frequently suffer from issues like low-quality waveforms and low text-audio consistency, hindering the advancement of music generation models. To address these challenges, we propose a novel quality-aware training paradigm for generating high-quality, high-musicality music from large-scale, quality-imbalanced datasets. Additionally, by leveraging unique properties in the latent space of musical signals, we adapt and implement a masked diffusion transformer (MDT) model for the TTM task, showcasing its capacity for quality control and enhanced musicality. Furthermore, we introduce a three-stage caption refinement approach to address low-quality captions' issue. Experiments show state-of-the-art (SOTA) performance on benchmark datasets including MusicCaps and the Song-Describer Dataset with both objective and subjective metrics. Demo audio samples are available at https://qa-mdt.github.io/, code and pretrained checkpoints are open-sourced at https://github.com/ivcylc/OpenMusic.

cs.SD↗

arXiv2025

Spin Grouping in Ring Cavity and its Protection on Entangled States Transfer

Long-range interactions are essential for large-scale quantum computation and quantum interconnections. Cavities provide a promising avenue to achieve long-range interaction by enhancing the coupling of remote qubits through shared cavity modes. In this work, we investigate a spin array coupled to a ring cavity supporting two counterpropagating modes, focusing on the system's eigenstates and spin dynamics in the low-excitation regime. We show that, under specific spatial configurations, the spins naturally self-organize into two groups, within which exciton transport is confined. This spin-grouping mechanism preserves coherence between spins across the two groups, and is leveraged to deterministically transfer entangled states between remote spin pairs with additional dynamical addressing. We further propose feasible implementations using atomic qubits or solid-state platforms. Our scheme enables entangling remote spins within a cavity, highlighting potential applications in scalable quantum information processing.

quant-ph↗

arXiv2025

UDHF2-Net: Uncertainty-diffusion-model-based High-Frequency TransFormer Network for Remotely Sensed Imagery Interpretation

Remotely sensed imagery interpretation (RSII) faces the three major problems: (1) objective representation of spatial distribution patterns; (2) edge uncertainty problem caused by downsampling encoder and intrinsic edge noises (e.g., mixed pixel and edge occlusion etc.); and (3) false detection problem caused by geometric registration error in change detection. To solve the aforementioned problems, uncertainty-diffusion-model-based high-Frequency TransFormer network (UDHF2-Net) is the first to be proposed, whose superiorities are as follows: (1) a spatially-stationary-and-non-stationary high-frequency connection paradigm (SHCP) is proposed to enhance the interaction of spatially frequency-wise stationary and non-stationary features to yield high-fidelity edge extraction result. Inspired by HRFormer, SHCP proposes high-frequency-wise stream to replace high-resolution-wise stream in HRFormer through the whole encoder-decoder process with parallel frequency-wise high-to-low streams, so it improves the edge extraction accuracy by continuously remaining high-frequency information; (2) a mask-and-geo-knowledge-based uncertainty diffusion module (MUDM), which is a self-supervised learning strategy, is proposed to improve the edge accuracy of extraction and change detection by gradually removing the simulated spectrum noises based on geo-knowledge and the generated diffused spectrum noises; (3) a frequency-wise semi-pseudo-Siamese UDHF2-Net is the first to be proposed to balance accuracy and complexity for change detection. Besides the aforementioned spectrum noises in semantic segmentation, MUDM is also a self-supervised learning strategy to effectively reduce the edge false change detection from the generated imagery with geometric registration error.

cs.CV↗

arXiv2025

Infinite combinatorial Ricci flow in spherical background geometry

Since the fundamental work of Chow-Luo \cite{CL03}, Ge \cite{Ge12,Ge17} et al., the combinatorial curvature flow methods became a basic technique in the study of circle pattern theory. In this paper, we investigate the combinatorial Ricci flow with prescribed total geodesic curvatures in spherical background geometry. For infinite cellular decompositions, we establish the existence of a solution to the flow equation for all time. Furthermore, under an additional condition, we prove that the solution converges as time tends to infinity. To the best of our knowledge, this is the first study of an infinite combinatorial curvature flow in spherical background geometry.

math.GT↗

arXiv2025

Tailoring composite skyrmionic spin textures in an above-room-temperature ferromagnet Fe3-xGaTe2

Realizing room-temperature tunable skyrmionic objects in van der Waals ferromagnet offers unparalleled prospects for future spintronics. Here, we report an experimental investigation on the emergence and evolution of skyrmionic spin textures in the non-stoichiometric Fe3-xGaTe2 using magnetic force microscopy. The iron-deficiency-specific magnetic states of stripe, striped skyrmionium and striped skyrmion sack are observed. Through zero-field-cooling and field-cooling measurements, we observed distinct topological transitions and trivial transitions (distinguished by changes in topological charge) emerging during the stepwise evolution of topological spin textures, which enabled us to develop an evolution pathway model. Leveraging this model, the room-temperature stable composite topological spin textures of skyrmionium, skyrmion bag and sack states are further controllably realized via the exclusive topological-transition path (regulated by magnetic field and DMI intensity). Our work provides valuable insights into the room-temperature realization of topological spin textures in Fe3-xGaTe2, and inspires further exploration of their potential applications in heterostructure spintronics.

cond-mat.mtrl-sci↗

arXiv2025

FLSSM: A Federated Learning Storage Security Model with Homomorphic Encryption

Federated learning based on homomorphic encryption has received widespread attention due to its high security and enhanced protection of user data privacy. However, the characteristics of encrypted computation lead to three challenging problems: ``computation-efficiency", ``attack-tracing" and ``contribution-assessment". The first refers to the efficiency of encrypted computation during model aggregation, the second refers to tracing malicious attacks in an encrypted state, and the third refers to the fairness of contribution assessment for local models after encryption. This paper proposes a federated learning storage security model with homomorphic encryption (FLSSM) to protect federated learning model privacy and address the three issues mentioned above. First, we utilize different nodes to aggregate local models in parallel, thereby improving encrypted models' aggregation efficiency. Second, we introduce trusted supervise nodes to examine local models when the global model is attacked, enabling the tracing of malicious attacks under homomorphic encryption. Finally, we fairly reward local training nodes with encrypted local models based on trusted training time. Experiments on multiple real-world datasets show that our model significantly outperforms baseline models in terms of both efficiency and security metrics.

cs.CR↗

arXiv2025

Analytical Modeling of Real-World Photonic Quantum Teleportation

We develop analytical models for realistic photonic quantum teleportation experiments with time-bin qubits, utilizing phase space methods from quantum optics. These models yield analytical expressions for Hong-Ou-Mandel interference visibilities and qubit fidelities, accounting for imperfections such as loss, photon distinguishability, and unwanted multi-photon events in single and entangled photon sources. Our expressions agree with the Hong-Ou-Mandel interference visibilities and teleportation fidelities reported by Valivarthi et al. (PRX Quantum 1, 020317, 2020). We use these models to predict and analyze the outcomes of future teleportation experiments under varying degrees of imperfections.

quant-ph↗

arXiv2025

An Atomic Skill Library Construction Method for Data-Efficient Embodied Manipulation

Embodied manipulation is a fundamental ability in the realm of embodied artificial intelligence. Although current embodied manipulation models show certain generalizations in specific settings, they struggle in new environments and tasks due to the complexity and diversity of real-world scenarios. The traditional end-to-end data collection and training manner leads to significant data demands. Decomposing end-to-end tasks into atomic skills helps reduce data requirements and improves the task success rate. However, existing methods are limited by predefined skill sets that cannot be dynamically updated. To address the issue, we introduce a three-wheeled data-driven method to build an atomic skill library. We divide tasks into subtasks using the Vision-Language-Planning (VLP). Then, atomic skill definitions are formed by abstracting the subtasks. Finally, an atomic skill library is constructed via data collection and Vision-Language-Action (VLA) fine-tuning. As the atomic skill library expands dynamically with the three-wheel update strategy, the range of tasks it can cover grows naturally. In this way, our method shifts focus from end-to-end tasks to atomic skills, significantly reducing data costs while maintaining high performance and enabling efficient adaptation to new tasks. Extensive experiments in real-world settings demonstrate the effectiveness and efficiency of our approach.

cs.RO↗

arXiv2025

Bridge-SR: Schrödinger Bridge for Efficient SR

Speech super-resolution (SR), which generates a waveform at a higher sampling rate from its low-resolution version, is a long-standing critical task in speech restoration. Previous works have explored speech SR in different data spaces, but these methods either require additional compression networks or exhibit limited synthesis quality and inference speed. Motivated by recent advances in probabilistic generative models, we present Bridge-SR, a novel and efficient any-to-48kHz SR system in the speech waveform domain. Using tractable Schrödinger Bridge models, we leverage the observed low-resolution waveform as a prior, which is intrinsically informative for the high-resolution target. By optimizing a lightweight network to learn the score functions from the prior to the target, we achieve efficient waveform SR through a data-to-data generation process that fully exploits the instructive content contained in the low-resolution observation. Furthermore, we identify the importance of the noise schedule, data scaling, and auxiliary loss functions, which further improve the SR quality of bridge-based systems. The experiments conducted on the benchmark dataset VCTK demonstrate the efficiency of our system: (1) in terms of sample quality, Bridge-SR outperforms several strong baseline methods under different SR settings, using a lightweight network backbone (1.7M); (2) in terms of inference speed, our 4-step synthesis achieves better performance than the 8-step conditional diffusion counterpart (LSD: 0.911 vs 0.927). Demo at https://bridge-sr.github.io.

cs.SD↗

arXiv2024

ΩSFormer: Dual-Modal Ω-like Super-Resolution Transformer Network for Cross-scale and High-accuracy Terraced Field Vectorization Extraction

Terraced field is a significant engineering practice for soil and water conservation (SWC). Terraced field extraction from remotely sensed imagery is the foundation for monitoring and evaluating SWC. This study is the first to propose a novel dual-modal Ω-like super-resolution Transformer network for intelligent TFVE, offering the following advantages: (1) reducing edge segmentation error from conventional multi-scale downsampling encoder, through fusing original high-resolution features with downsampling features at each step of encoder and leveraging a multi-head attention mechanism; (2) improving the accuracy of TFVE by proposing a Ω-like network structure, which fully integrates rich high-level features from both spectral and terrain data to form cross-scale super-resolution features; (3) validating an optimal fusion scheme for cross-modal and cross-scale (i.e., inconsistent spatial resolution between remotely sensed imagery and DEM) super-resolution feature extraction; (4) mitigating uncertainty between segmentation edge pixels by a coarse-to-fine and spatial topological semantic relationship optimization (STSRO) segmentation strategy; (5) leveraging contour vibration neural network to continuously optimize parameters and iteratively vectorize terraced fields from semantic segmentation results. Moreover, a DMRVD for deep-learning-based TFVE was created for the first time, which covers nine study areas in four provinces of China, with a total coverage area of 22441 square kilometers. To assess the performance of ΩSFormer, classic and SOTA networks were compared. The mIOU of ΩSFormer has improved by 0.165, 0.297 and 0.128 respectively, when compared with best accuracy single-modal remotely sensed imagery, single-modal DEM and dual-modal result.

cs.CV↗

arXiv2024

DBF-Net: A Dual-Branch Network with Feature Fusion for Ultrasound Image Segmentation

Accurately segmenting lesions in ultrasound images is challenging due to the difficulty in distinguishing boundaries between lesions and surrounding tissues. While deep learning has improved segmentation accuracy, there is limited focus on boundary quality and its relationship with body structures. To address this, we introduce UBBS-Net, a dual-branch deep neural network that learns the relationship between body and boundary for improved segmentation. We also propose a feature fusion module to integrate body and boundary information. Evaluated on three public datasets, UBBS-Net outperforms existing methods, achieving Dice Similarity Coefficients of 81.05% for breast cancer, 76.41% for brachial plexus nerves, and 87.75% for infantile hemangioma segmentation. Our results demonstrate the effectiveness of UBBS-Net for ultrasound image segmentation. The code is available at https://github.com/apple1986/DBF-Net.

eess.IV↗

arXiv2024

Compact optical waveform generator with digital feedback

A key requirement for quantum technologies based on atoms, ions, and molecules, is the ability to realize precise phase- and amplitude-controlled quantum operations via coherent laser pulses. However, for generating pulses on the sub-microsecond timescale, the characteristics of the optical and electronic components can introduce unwanted distortions that have a detrimental effect on the fidelity of quantum operations. In this paper, we present a compact arbitrary waveform generator that integrates a double-pass acousto-optic modulator for user-specified laser amplitude and phase modulations. Additionally, the module integrates an optical heterodyne detector to extract the precise laser pulse shape in real-time. The measured pulse shape is then fed into a digital feedback loop used to estimate the complex-valued transfer function and pre-distorted input pulses. We demonstrate the performance by generating shaped laser pulses suitable for realizing quantum logic gates with durations down to 180\,ns, requiring only a small number of feedback iterations.

quant-ph↗