arXiv subjects

Xing Deng

Publications and source records attributed to Xing Deng.

14 recordsLinked to original sources

arXiv2026

When Semantics Saturate or Emerge: Adaptation-Conditional Semantic Utility in Source-Free Cross-Domain Few-Shot Learning

Language descriptions in source-free cross-domain few-shot learning (SF-CDFSL) are often selected according to zero-shot accuracy obtained with a frozen vision--language model. This paper asks whether that ranking remains valid after target-domain visual adaptation. Under a strictly paired protocol, we compare a generic class-name template with fixed detailed class descriptions before and after visual Low-Rank Adaptation (LoRA) on EuroSAT, CropDisease, ISIC, and ChestX. Let $\deltazero$ and $\deltalora$ denote the Detailed-minus-Base accuracy before and after adaptation, respectively. Two recurring regimes emerge. In \emph{semantic saturation}, $\deltazero>0$ but $0<\deltalora\ll\deltazero$: on EuroSAT and CropDisease, initial gains of 8.13--21.54 percentage points contract to 0.69--2.96 points after LoRA. In \emph{semantic emergence}, $\deltazero\leq0$ but $\deltalora>0$: on ISIC and ChestX, detailed descriptions become more useful only after the visual representation is updated. Training trajectories and sample-level decomposition show that saturation is driven mainly by Base-LoRA recovering errors already solved by detailed semantics, whereas emergence is associated with prediction turnover and newly formed Detailed-only correct decisions. Fixed-point-free shuffled-semantic controls, a second CLIP backbone, and multiple random seeds support the broad pattern while identifying ChestX 1-shot as a weak boundary case. These findings establish that zero-shot prompt quality is an incomplete proxy for adaptation-anchor quality and motivate evaluating language on both sides of the adaptation boundary.

cs.CV↗

arXiv2026

LieTrunc-QNN: Lie Algebra Truncation and Quantum Expressivity Phase Transition from LiePrune to Provably Stable Quantum Neural Networks

Quantum Machine Learning (QML) is fundamentally limited by two challenges: barren plateaus (exponentially vanishing gradients) and the fragility of parameterized quantum circuits under noise. Despite extensive empirical studies, a unified theoretical framework remains lacking. We introduce LieTrunc-QNN, an algebraic-geometric framework that characterizes trainability via Lie-generated dynamics. Parameterized quantum circuits are modeled as Lie subalgebras of u(2^n), whose action induces a Riemannian manifold of reachable quantum states. Expressivity is reinterpreted as intrinsic manifold dimension and geometry. We establish a geometric capacity-plateau principle: increasing effective dimension leads to exponential gradient suppression due to concentration of measure. By restricting to structured Lie subalgebras (LieTrunc), the manifold is contracted, preventing concentration and preserving non-degenerate gradients. We prove two main results: (1) a trainability lower bound for LieTrunc-QNN, and (2) that the Fubini-Study metric rank is bounded by the algebraic span of generators, showing expressivity is governed by structure rather than parameter count. Compact Lie subalgebras also provide inherent robustness to perturbations. Importantly, we establish a polynomial trainability regime where gradient variance decays polynomially instead of exponentially. Experiments (n=2-6) validate the theory: LieTrunc-QNN maintains stable gradients and high effective dimension, while random truncation leads to metric rank collapse. At n=6, full metric rank is preserved (rank=16). Results support a scaling law between gradient variance and effective dimension. This work provides a unified geometric framework for QNN design, linking Lie algebra, manifold geometry, and optimization.

cs.LG↗

arXiv2026

Enhancing Imbalanced Electrocardiogram Classification: A Novel Approach Integrating Data Augmentation through Wavelet Transform and Interclass Fusion

Imbalanced electrocardiogram (ECG) data hampers the efficacy and resilience of algorithms in the automated processing and interpretation of cardiovascular diagnostic information, which in turn impedes deep learning-based ECG classification. Notably, certain cardiac conditions that are infrequently encountered are disproportionately underrepresented in these datasets. Although algorithmic generation and oversampling of specific ECG signal types can mitigate class skew, there is a lack of consensus regarding the effectiveness of such techniques in ECG classification. Furthermore, the methodologies and scenarios of ECG acquisition introduce noise, further complicating the processing of ECG data. This paper presents a significantly enhanced ECG classifier that simultaneously addresses both class imbalance and noise-related challenges in ECG analysis, as observed in the CPSC 2018 dataset. Specifically, we propose the application of feature fusion based on the wavelet transform, with a focus on wavelet transform-based interclass fusion, to generate the training feature library and the test set feature library. Subsequently, the original training and test data are amalgamated with their respective feature databases, resulting in more balanced training and test datasets. Employing this approach, our ECG model achieves recognition accuracies of up to 99%, 98%, 97%, 98%, 96%, 92%, and 93% for Normal, AF, I-AVB, LBBB, RBBB, PAC, PVC, STD, and STE, respectively. Furthermore, the average recognition accuracy for these categories ranges between 92\% and 98\%. Notably, our proposed data fusion methodology surpasses any known algorithms in terms of ECG classification accuracy in the CPSC 2018 dataset.

cs.LG↗

arXiv2026

Spectral Complex Autoencoder Pruning: A Fidelity-Guided Criterion for Extreme Structured Channel Compression

We propose Spectral Complex Autoencoder Pruning (SCAP), a reconstruction-based criterion that measures functional redundancy at the level of individual output channels. For each convolutional layer, we construct a complex interaction field by pairing the full multi-channel input activation as the real part with a single output-channel activation (spatially aligned and broadcast across input channels) as the imaginary part. We transform this complex field to the frequency domain and train a low-capacity autoencoder to reconstruct normalized spectra. Channels whose spectra are reconstructed with high fidelity are interpreted as lying close to a low-dimensional manifold captured by the autoencoder and are therefore more compressible; conversely, channels with low fidelity are retained as they encode information that cannot be compactly represented by the learned manifold. This yields an importance score (optionally fused with the filter L1 norm) that supports simple threshold-based pruning and produces a structurally consistent pruned network. On VGG16 trained on CIFAR-10, at a fixed threshold of 0.6, we obtain 90.11% FLOP reduction and 96.30% parameter reduction with an absolute Top-1 accuracy drop of 1.67% from a 93.44% baseline after fine-tuning, demonstrating that spectral reconstruction fidelity of complex interaction fields is an effective proxy for channel-level redundancy under aggressive compression.

cs.CV↗

arXiv2025

One-Shot Structured Pruning of Quantum Neural Networks via $q$-Group Engineering and Quantum Geometric Metrics

Quantum neural networks (QNNs) suffer from severe gate-level redundancy, which hinders their deployment on noisy intermediate-scale quantum (NISQ) devices. In this work, we propose q-iPrune, a one-shot structured pruning framework grounded in the algebraic structure of $q$-deformed groups and task-conditioned quantum geometry. Unlike prior heuristic or gradient-based pruning methods, q-iPrune formulates redundancy directly at the gate level. Each gate is compared within an algebraically consistent subgroup using a task-conditioned $q$-overlap distance, which measures functional similarity through state overlaps on a task-relevant ensemble. A gate is removed only when its replacement by a subgroup representative provably induces a bounded deviation on all task observables. We establish three rigorous theoretical guarantees. First, we prove completeness of redundancy pruning: no gate that violates the prescribed similarity threshold is removed. Second, we show that the pruned circuit is functionally equivalent up to an explicit, task-conditioned error bound, with a closed-form dependence on the redundancy tolerance and the number of replaced gates. Third, we prove that the pruning procedure is computationally feasible, requiring only polynomial-time comparisons and avoiding exponential enumeration over the Hilbert space. To adapt pruning decisions to hardware imperfections, we introduce a noise-calibrated deformation parameter $\lambda$ that modulates the $q$-geometry and redundancy tolerance. Experiments on standard quantum machine learning benchmarks demonstrate that q-iPrune achieves substantial gate reduction while maintaining bounded task performance degradation, consistent with our theoretical guarantees.

quant-ph↗

arXiv2025

LiePrune: Lie Group and Quantum Geometric Dual Representation for One-Shot Structured Pruning of Quantum Neural Networks

Quantum neural networks (QNNs) and parameterized quantum circuits (PQCs) are key building blocks for near-term quantum machine learning. However, their scalability is constrained by excessive parameters, barren plateaus, and hardware limitations. We propose LiePrune, the first mathematically grounded one-shot structured pruning framework for QNNs that leverages Lie group structure and quantum geometric information. Each gate is jointly represented in a Lie group--Lie algebra dual space and a quantum geometric feature space, enabling principled redundancy detection and aggressive compression. Experiments on quantum classification (MNIST, FashionMNIST), quantum generative modeling (Bars-and-Stripes), and quantum chemistry (LiH VQE) show that LiePrune achieves over $10\times$ compression with negligible or even improved task performance, while providing provable guarantees on redundancy detection, functional approximation, and computational complexity.

quant-ph↗

arXiv2025

Quantum-Inspired Spectral Geometry for Neural Operator Equivalence and Structured Pruning

The rapid growth of multimodal intelligence on resource-constrained and heterogeneous domestic hardware exposes critical bottlenecks: multimodal feature heterogeneity, real-time requirements in dynamic scenarios, and hardware-specific operator redundancy. This work introduces a quantum-inspired geometric framework for neural operators that represents each operator by its normalized singular value spectrum on the Bloch hypersphere. We prove a tight spectral-to-functional equivalence theorem showing that vanishing Fubini--Study/Wasserstein-2 distance implies provable functional closeness, establishing the first rigorous foundation for cross-modal and cross-architecture operator substitutability. Based on this metric, we propose Quantum Metric-Driven Functional Redundancy Graphs (QM-FRG) and one-shot structured pruning. Controlled simulation validates the superiority of the proposed metric over magnitude and random baselines. An extensive experimental validation on large-scale multimodal transformers and domestic heterogeneous hardware (Huawei Ascend, Cambricon MLU, Kunlunxin) hardware is deferred to an extended journal version currently in preparation.

cs.CV↗

arXiv2023

Fragment and Integrate Network (FIN): A Novel Spatial-Temporal Modeling Based on Long Sequential Behavior for Online Food Ordering Click-Through Rate Prediction

Spatial-temporal information has been proven to be of great significance for click-through rate prediction tasks in online Location-Based Services (LBS), especially in mainstream food ordering platforms such as DoorDash, Uber Eats, Meituan, and Ele.me. Modeling user spatial-temporal preferences with sequential behavior data has become a hot topic in recommendation systems and online advertising. However, most of existing methods either lack the representation of rich spatial-temporal information or only handle user behaviors with limited length, e.g. 100. In this paper, we tackle these problems by designing a new spatial-temporal modeling paradigm named Fragment and Integrate Network (FIN). FIN consists of two networks: (i) Fragment Network (FN) extracts Multiple Sub-Sequences (MSS) from lifelong sequential behavior data, and captures the specific spatial-temporal representation by modeling each MSS respectively. Here both a simplified attention and a complicated attention are adopted to balance the performance gain and resource consumption. (ii) Integrate Network (IN) builds a new integrated sequence by utilizing spatial-temporal interaction on MSS and captures the comprehensive spatial-temporal representation by modeling the integrated sequence with a complicated attention. Both public datasets and production datasets have demonstrated the accuracy and scalability of FIN. Since 2022, FIN has been fully deployed in the recommendation advertising system of Ele.me, one of the most popular online food ordering platforms in China, obtaining 5.7% improvement on Click-Through Rate (CTR) and 7.3% increase on Revenue Per Mille (RPM).

cs.IR↗

arXiv2022

Giant superlinear power dependence of photocurrent based on layered Ta$_2$NiS$_5$ photodetector

Photodetector based on two-dimensional (2D) materials is an ongoing quest in optoelectronics. These 2D photodetectors are generally efficient at low illuminating power but suffer severe recombination processes at high power, which results in the sublinear power dependence of photoresponse and lower optoelectronic efficiency. The desirable superlinear photocurrent is mostly achieved by sophisticated 2D heterostructures or device arrays, while 2D materials rarely show intrinsic superlinear photoresponse. Here, we report the giant superlinear power dependence of photocurrent based on multi-layer Ta$_2$NiS$_5$. While the fabricated photodetector exhibits good sensitivity ($3.1 mS/W$ per square) and fast photoresponse ($31 \mu$$s$), the bias-, polarization-, and spatial-resolved measurements point to an intrinsic photoconductive mechanism. By increasing the incident power density from $1.5 \mu$W/$\mu$$m^{2}$ to $200 \mu$W/$\mu$$m^{2}$, the photocurrent power dependence varies from sublinear to superlinear. At higher illuminating conditions, a prominent superlinearity is observed with a giant power exponent of $\gamma=1.5$. The unusual photoresponse can be explained by a two-recombination-center model where the distinct density of states of the recombination centers effectively closes all recombination channels. The fabricated photodetector is integrated into camera for taking photos with enhanced contrast due to the superlinearity. Our work provides an effective route to enable higher optoelectronic efficiency at extreme conditions.

physics.app-ph↗

arXiv2022

Manifold formation and crossings of ultracold lattice spinor atoms in the intermediate interaction regime

Ultracold spinor atoms in the weak and strong interaction regime have received extensive investigations, while the behavior in the intermediate regime is less understood. We numerically investigate ultracold spinor atomic ensembles of finite size in the intermediate interaction regime, and reveal the evolution of the eigenstates from the strong to the intermediate regime. In the strong interaction regime, it has been well known that the low-lying eigenenergy spectrum presents the well-gaped multi-manifold structure, and the energy gaps protect the categorization of the eigenstates. In the intermediate interaction regime, it is found that the categorization of the eigenstates is preserved, and the eigenenergy spectrum become quasi-continuum, with different manifolds becoming overlapped. The overlapping induces both direct and avoided crossings between close-lying manifolds, which is determined by the combined symmetries of the eigenstates involved in the crossing. A modified t-J model is derived to describe the low-lying eigenstates in the intermediate regime, which can capture the formation and crossings of the manifolds. State preparation through the avoided crossings is also investigated.

cond-mat.quant-gas↗

arXiv2021

Interaction effects of pseudospin-based magnetic monopoles and kinks in a doped dipolar superlattice gas

Magnetic monopoles and kinks are topological excitations extensively investigated in quantum spin systems, but usually they are studied in different setups. We explore the conditions for the coexistence and the interaction effects of these quasiparticles in the pseudospin chain of the atomic dipolar superlattice gas. In this chain, the magnetic kink is the intrinsic quasiparticle, and the particle/hole defect takes over the role of the north/south magnetic monopole, exerting monopolar magnetic fields to neighboring spins. A confinement effect between the monopole and kink is revealed, which renormalizes the dispersion of the kink. The corresponding dynamical deconfinement process is observed and arises due to the kink-antikink annihilation. The rich interaction effects of the two quasiparticles could stimulate corresponding investigations in bulk spin systems.

cond-mat.quant-gas↗

arXiv2021

Quantum metrology with precision reaching beyond-$1/N$ scaling through $N$-probe entanglement generating interactions

Nonlinear interactions are recognized as potential resources for quantum metrology, facilitating parameter estimation precisions that scale as the exponential Heisenberg limit of $2^{-N}$. We explore such nonlinearity and propose an associated quantum measurement scenario based on the nonlinear interaction of $N$-probe entanglement generating form. This scenario provides an enhanced precision scaling of $D^{-N}/(N-1)!$ with $D > 2$ a tunable parameter. In addition, it can be readily implemented in a variety of experimental platforms and applied to measurements of a wide range of quantities, including local gravitational acceleration $g$, magnetic field, and its higher-order gradients.

quant-ph↗

arXiv2017

Generating scalable entanglement of ultracold bosons in superlattices through resonant shaking

Based on a one-dimensional double-well superlattice with a unit filling of ultracold atoms per site, we propose a scheme to generate scalable entangled states in the superlattice through resonant lattice shakings. Our scheme utilizes periodic lattice modulations to entangle two atoms in each unit cell with respect to their orbital degree of freedom, and the complete atomic system in the superlattice becomes a cluster of bipartite entangled atom pairs. To demonstrate this we perform $ab \ initio$ quantum dynamical simulations using the Multi-Layer Multi-Configuration Time-Dependent Hartree Method for Bosons, which accounts for all correlations among the atoms. The proposed clusters of bipartite entanglements manifest as an essential resource for various quantum applications, such as measurement based quantum computation. The lattice shaking scheme to generate this cluster possesses advantages such as a high scalability, fast processing speed, rich controllability on the target entangled states, and accessibility with current experimental techniques.

cond-mat.quant-gas↗

arXiv2016

Collective excitations of dipolar gases based on local tunneling in superlattices

The collective dynamics of a dipolar fermionic quantum gas confined in a one-dimensional double-well superlattice is explored. The fermionic gas resides in a paramagnetic-like ground state in the weak interaction regime, upon which a new type of collective dynamics is found when applying a local perturbation. This dynamics is composed of the local tunneling of fermions in separate supercells, and is a pure quantum effect, with no classical counterpart. Due to the presence of the dipolar interactions the local tunneling transports through the entire superlattice, giving rise to a collective dynamics. A well-defined momentum-energy dispersion relation is identified in the ab-initio simulations demonstrating the phonon-like behavior. The phonon-like characteristic is also confirmed by an analytical description of the dynamics within a semiclassical picture.

cond-mat.quant-gas↗