arXiv subjects

Xianchao Zhang

Publications and source records attributed to Xianchao Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Joint Optimization for Federated Learning and Transmission over Unreliable Wireless Networks with Heterogeneous Data

In wireless federated learning (FL), data heterogeneity and multiple local updates induce client drift, degrading model convergence. It is further affected by unreliable wireless links, as transmission errors may invalidate model updates. To address these challenges, we propose a federated random walk averaging (FedRW) framework, which is a variant of federated averaging (FedAvg) that mitigates data heterogeneity by updating models along random walk (RW) paths and aggregating them at the server. Model parameters are transmitted in packets with retransmission support to improve training quality by mitigating wireless errors along RW paths. Meanwhile, wireless transmission delays hinder the exploration of FedRW. To this end, we formulate a joint optimization problem that integrates learning, RW path selection, and transmission parameter tuning, aiming to minimize the training loss under delay constraints. By deriving an upper bound on the expected convergence of FedRW over unreliable wireless networks, we reduce the problem to a general form agnostic to task type and model architecture. A distributed solution is then proposed, in which the server or clients optimize packet size and maximum number of retransmissions locally, and efficiently select reliable and expandable next-hop nodes via a resilience-aware beam search with dynamic pruning. Simulation results show that FedRW achieves 2.26%-9% higher accuracy than state-of-the-art baselines under high data heterogeneity. Furthermore, the jointly optimized FedRW yields at least 2.78% higher accuracy and faster convergence compared to baselines.

cs.DC↗

arXiv2026

GenSpan: Generation-Calibrated Motion Span Priors for Multi-Verb Video Corpus Moment Retrieval

Video Corpus Moment Retrieval (VCMR) aims to retrieve both the correct video and its temporal segment corresponding to a natural-language query, a task that is especially challenging for multi-verb queries where temporal action ordering is critical. Existing approaches often rely solely on text or static images and struggle to capture implicit motion dynamics, leading to retrieval errors and temporal misalignment. We propose GenSpan, a generation-calibrated VCMR framework that constructs short auxiliary videos from LLM-selected subtitle cues and decomposed sub-events, using these as temporal priors rather than direct retrieval targets. A token selector filters candidate-video features aligned with generated motion, and a bidirectional state-space model efficiently predicts video-moment tuples. Experiments on TVR and ActivityNet-Captions demonstrate that GenSpan improves corpus-level retrieval and moment localization, particularly for complex multi-action queries, while reducing computational cost compared to state-of-the-art multimodal baselines.

cs.CV↗

arXiv2026

Hybrid RIS-Aided Digital Over-the-Air Computing for Edge AI Inference: Joint Feature Quantization and Active-Passive Beamforming Design

The vision of 6G networks aims to enable edge inference by leveraging ubiquitously deployed artificial intelligence (AI) models, facilitating intelligent environmental perception for a wide range of applications. A critical operation in edge inference is for an edge node (EN) to aggregate multi-view sensory features extracted by distributed agents, thereby boosting perception accuracy. Over-the-air computing (AirComp) emerges as a promising technique for rapid feature aggregation by exploiting the waveform superposition property of analog-modulated signals, which is, however, incompatible with existing digital communication systems. Meanwhile, hybrid reconfigurable intelligent surface (RIS), a novel RIS architecture capable of simultaneous signal amplification and reflection, exhibits potential for enhancing AirComp. Therefore, this paper proposes a Hybrid RIS-aided Digital AirComp (HRD-AirComp) scheme, which employs vector quantization to map high-dimensional features into discrete codewords that are digitally modulated into symbols for wireless transmission. By judiciously adjusting the AirComp transceivers and hybrid RIS reflection to control signal superposition across agents, the EN can estimate the aggregated features from the received signals. To endow HRD-AirComp with a task-oriented design principle, we derive a surrogate function for inference accuracy that characterizes the impact of feature quantization and over-the-air aggregation. Based on this surrogate, we formulate an optimization problem targeting inference accuracy maximization, and develop an efficient algorithm to jointly optimize the quantization bit allocation, agent transmission coefficients, EN receiving beamforming, and hybrid RIS reflection beamforming. Experimental results demonstrate that the proposed HRD-AirComp outperforms baselines in terms of both inference accuracy and uncertainty.

cs.NI↗

arXiv2025

LISA: A Layer-wise Integration and Suppression Approach for Hallucination Mitigation in Multimodal Large Language Models

Multimodal Large Language Models (MLLMs) excel in vision-language tasks such as image captioning but remain prone to object hallucinations, where they describe objects that do not appear in the image. To mitigate this, we propose LISA, a Layer-wise Integration and Suppression Approach. LISA leverages the layer-wise functional roles in MLLMs: shallow layers provide visual grounding, middle layers encode semantics, and deep layers tend to amplify spurious signals. First, layer-wise spectral modulation stabilizes attention by suppressing over-amplified activations in deeper layers while preserving alignment cues in earlier layers. Second, token-level logits from selected layers are fused via anchor-based routing, with token-wise anchor selection and soft logit fusion enabling adaptive integration during decoding. LISA is fully plug-and-play and can be seamlessly integrated into existing MLLMs, including Qwen2.5-VL. Experiments on multiple benchmarks show that LISA reduces hallucinations by up to 53.6% in $\text{CHAIR}_\text{I}$ and improves POPE F1 by up to 5.1%, demonstrating strong generalization across models and tasks. Our code is available at https://github.com/zhlisa1010-eng/LISA.

cs.CV↗

arXiv2025

Bose-Einstein condensates in a spin-twisted harmonic trap

We investigate the ground-state phases and spin-scissors dynamics of binary Bose-Einstein condensates confined in a twisted two-dimensional harmonic trap. The ground state hosts three distinct phases-phase-separated, polarized, and phase-mixed-determined by the Rabi coupling, interaction ratio G (between inter-component and intra-component interactions), and spin-twisting which induces edge-localized polarization through position-dependent detuning. In the phase-mixed regime, the ground state is characterized by a finite spin-scissors susceptibility and can be accurately described using local density approximation. In the dynamics, the system exhibits stable periodic beating in the phase-mixed state for G<1. For G>1, its evolution progresses from beat damping (phase-separated state) to polarized relaxation (polarized state), finally reaching stable periodic beating (phase-mixed state) after a finite waiting time. The dependence of the waiting time contrasts sharply with the monotonic behavior of one-dimensional spin-dipole dynamics, revealing qualitatively distinct mechanisms governed by geometry and interactions. In summary, these results establish a unified link between ground-state properties and nonequilibrium responses in twisted spinor condensates, offering a versatile platform for exploring spin-related quantum many-body phenomena.

cond-mat.quant-gas↗

arXiv2025

Image Super-Resolution-Based Signal Enhancement in Bistatic ISAC

Bistatic Integrated Sensing and Communication (ISAC) is poised to become a cornerstone technology in next-generation communication networks, such as Beyond 5G (B5G) and 6G, by enabling the concurrent execution of sensing and communication functions without requiring significant modifications to existing infrastructure. Despite its promising potential, a major challenge in bistatic cooperative sensing lies in the degradation of sensing accuracy, primarily caused by the inherently weak received signals resulting from high reflection losses in complex environments. Traditional methods have predominantly relied on adaptive filtering techniques to enhance the Signal-to-Noise Ratio (SNR) by dynamically adjusting the filter coefficients. However, these methods often struggle to adapt effectively to the increasingly complex and diverse network topologies. To address these challenges, we propose a novel Image Super-Resolution-based Signal Enhancement (ISR-SE) framework that significantly improves the recognition and recovery capabilities of ISAC signals. Specifically, we first perform a time-frequency analysis by applying the Short-Time Fourier Transform (STFT) to the received signals, generating spectrograms that capture the frequency, magnitude, and phase components. These components are then mapped into RGB images, where each channel represents one of the extracted features, enabling a more intuitive and informative visualization of the signal structure. To enhance these RGB images, we design an improved denoising network that combines the strengths of the UNet architecture and diffusion models. This hybrid architecture leverages UNet's multi-scale feature extraction and the generative capacity of diffusion models to perform effective image denoising, thereby improving the quality and clarity of signal representations under low-SNR conditions.

eess.SP↗

arXiv2025

On Analysis of Superimposed Pilot in Multi-User Massive MIMO with Massive Connectivity

The simultaneous transmission of numerous users presents substantial challenges due to the inherent trade-off between channel estimation and information transmission in multi-user multiple-input multiple-output (MIMO) system. In this paper, we explore the use of the superimposed pilot (SP) scheme to tackle the large transmitting users, where the number of users may exceed the coherent time. SP scheme incorporates both transmitted data and noise in the channel estimation process, which is significant different from the counterpart of RP scheme. We provide an in-depth analysis of the interaction between interference caused by channel estimation errors and noise. We then derive the explicit expression for the scaling law of the mutual information lower bound (MILB) in relation to the number of users and the levels of transmitted power. Besides, the optimal power allocation between pilots and data transmission is also derived analytically. The analytical results demonstrate that the SP scheme significantly improves performance compared to traditional RP scheme in our consider case. Numerical results are also presented to validate our theoretical derivations.

cs.IT↗

arXiv2024

Hawkes based Representation Learning for Reasoning over Scale-free Community-structured Temporal Knowledge Graphs

Temporal knowledge graph (TKG) reasoning has become a hot topic due to its great value in many practical tasks. The key to TKG reasoning is modeling the structural information and evolutional patterns of the TKGs. While great efforts have been devoted to TKG reasoning, the structural and evolutional characteristics of real-world networks have not been considered. In the aspect of structure, real-world networks usually exhibit clear community structure and scale-free (long-tailed distribution) properties. In the aspect of evolution, the impact of an event decays with the time elapsing. In this paper, we propose a novel TKG reasoning model called Hawkes process-based Evolutional Representation Learning Network (HERLN), which learns structural information and evolutional patterns of a TKG simultaneously, considering the characteristics of real-world networks: community structure, scale-free and temporal decaying. First, we find communities in the input TKG to make the encoding get more similar intra-community embeddings. Second, we design a Hawkes process-based relational graph convolutional network to cope with the event impact-decaying phenomenon. Third, we design a conditional decoding method to alleviate biases towards frequent entities caused by long-tailed distribution. Experimental results show that HERLN achieves significant improvements over the state-of-the-art models.

cs.SI↗

arXiv2024

6D Movable Antenna Enhanced Interference Mitigation for Cellular-Connected UAV Communications

Cellular-connected unmanned aerial vehicle (UAV) communications is an enabling technology to transmit control signaling or payload data for UAVs through cellular networks. Due to the line-of-sight (LoS) dominant air-to-ground channels, efficient interference mitigation is crucial to UAV communications, while the conventional fixed-position antenna (FPA) arrays have limited degrees of freedom (DoFs) to suppress the interference between the UAV and its non-associated co-channel base stations (BSs). To address this challenge, we propose in this letter a new approach by utilizing the six-dimensional movable antenna (6DMA) arrays to enhance the interference mitigation for the UAV. Specifically, we propose an efficient block coordinate descent (BCD) algorithm to iteratively optimize the antenna position vector (APV), array rotation vector (ARV), receive beamforming vector, and associated BS of the UAV to maximize its signal-to-interference-plus-noise ratio (SINR). Numerical results show that the proposed 6DMA enhanced cellular-connected UAV communication can significantly outperform that with the traditional FPA arrays and other benchmark schemes in terms of interference mitigation.

eess.SP↗

arXiv2024

Neural Probabilistic Logic Learning for Knowledge Graph Reasoning

Knowledge graph (KG) reasoning is a task that aims to predict unknown facts based on known factual samples. Reasoning methods can be divided into two categories: rule-based methods and KG-embedding based methods. The former possesses precise reasoning capabilities but finds it challenging to reason efficiently over large-scale knowledge graphs. While gaining the ability to reason over large-scale knowledge graphs, the latter sacrifices reasoning accuracy. This paper aims to design a reasoning framework called Neural Probabilistic Logic Learning(NPLL) that achieves accurate reasoning on knowledge graphs. Our approach introduces a scoring module that effectively enhances the expressive power of embedding networks, striking a balance between model simplicity and reasoning capabilities. We improve the interpretability of the model by incorporating a Markov Logic Network based on variational inference. We empirically evaluate our approach on several benchmark datasets, and the experimental results validate that our method substantially enhances the accuracy and quality of the reasoning results.

cs.AI↗

arXiv2024

Liberating Seen Classes: Boosting Few-Shot and Zero-Shot Text Classification via Anchor Generation and Classification Reframing

Few-shot and zero-shot text classification aim to recognize samples from novel classes with limited labeled samples or no labeled samples at all. While prevailing methods have shown promising performance via transferring knowledge from seen classes to unseen classes, they are still limited by (1) Inherent dissimilarities among classes make the transformation of features learned from seen classes to unseen classes both difficult and inefficient. (2) Rare labeled novel samples usually cannot provide enough supervision signals to enable the model to adjust from the source distribution to the target distribution, especially for complicated scenarios. To alleviate the above issues, we propose a simple and effective strategy for few-shot and zero-shot text classification. We aim to liberate the model from the confines of seen classes, thereby enabling it to predict unseen categories without the necessity of training on seen classes. Specifically, for mining more related unseen category knowledge, we utilize a large pre-trained language model to generate pseudo novel samples, and select the most representative ones as category anchors. After that, we convert the multi-class classification task into a binary classification task and use the similarities of query-anchor pairs for prediction to fully leverage the limited supervision signals. Extensive experiments on six widely used public datasets show that our proposed method can outperform other strong baselines significantly in few-shot and zero-shot tasks, even without using any seen class samples.

cs.CV↗

arXiv2024

HQA-Attack: Toward High Quality Black-Box Hard-Label Adversarial Attack on Text

Black-box hard-label adversarial attack on text is a practical and challenging task, as the text data space is inherently discrete and non-differentiable, and only the predicted label is accessible. Research on this problem is still in the embryonic stage and only a few methods are available. Nevertheless, existing methods rely on the complex heuristic algorithm or unreliable gradient estimation strategy, which probably fall into the local optimum and inevitably consume numerous queries, thus are difficult to craft satisfactory adversarial examples with high semantic similarity and low perturbation rate in a limited query budget. To alleviate above issues, we propose a simple yet effective framework to generate high quality textual adversarial examples under the black-box hard-label attack scenarios, named HQA-Attack. Specifically, after initializing an adversarial example randomly, HQA-attack first constantly substitutes original words back as many as possible, thus shrinking the perturbation rate. Then it leverages the synonym set of the remaining changed words to further optimize the adversarial example with the direction which can improve the semantic similarity and satisfy the adversarial condition simultaneously. In addition, during the optimizing procedure, it searches a transition synonym word for each changed word, thus avoiding traversing the whole synonym set and reducing the query number to some extent. Extensive experimental results on five text classification datasets, three natural language inference datasets and two real-world APIs have shown that the proposed HQA-Attack method outperforms other strong baselines significantly.

cs.CL↗

arXiv2023

Boosting Decision-Based Black-Box Adversarial Attack with Gradient Priors

Decision-based methods have shown to be effective in black-box adversarial attacks, as they can obtain satisfactory performance and only require to access the final model prediction. Gradient estimation is a critical step in black-box adversarial attacks, as it will directly affect the query efficiency. Recent works have attempted to utilize gradient priors to facilitate score-based methods to obtain better results. However, these gradient priors still suffer from the edge gradient discrepancy issue and the successive iteration gradient direction issue, thus are difficult to simply extend to decision-based methods. In this paper, we propose a novel Decision-based Black-box Attack framework with Gradient Priors (DBA-GP), which seamlessly integrates the data-dependent gradient prior and time-dependent prior into the gradient estimation procedure. First, by leveraging the joint bilateral filter to deal with each random perturbation, DBA-GP can guarantee that the generated perturbations in edge locations are hardly smoothed, i.e., alleviating the edge gradient discrepancy, thus remaining the characteristics of the original image as much as possible. Second, by utilizing a new gradient updating strategy to automatically adjust the successive iteration gradient direction, DBA-GP can accelerate the convergence speed, thus improving the query efficiency. Extensive experiments have demonstrated that the proposed method outperforms other strong baselines significantly.

cs.LG↗

arXiv2023

Boosting Few-Shot Text Classification via Distribution Estimation

Distribution estimation has been demonstrated as one of the most effective approaches in dealing with few-shot image classification, as the low-level patterns and underlying representations can be easily transferred across different tasks in computer vision domain. However, directly applying this approach to few-shot text classification is challenging, since leveraging the statistics of known classes with sufficient samples to calibrate the distributions of novel classes may cause negative effects due to serious category difference in text domain. To alleviate this issue, we propose two simple yet effective strategies to estimate the distributions of the novel classes by utilizing unlabeled query samples, thus avoiding the potential negative transfer issue. Specifically, we first assume a class or sample follows the Gaussian distribution, and use the original support set and the nearest few query samples to estimate the corresponding mean and covariance. Then, we augment the labeled samples by sampling from the estimated distribution, which can provide sufficient supervision for training the classification model. Extensive experiments on eight few-shot text classification datasets show that the proposed method outperforms state-of-the-art baselines significantly.

cs.CL↗

arXiv2023

Infomaxformer: Maximum Entropy Transformer for Long Time-Series Forecasting Problem

The Transformer architecture yields state-of-the-art results in many tasks such as natural language processing (NLP) and computer vision (CV), since the ability to efficiently capture the precise long-range dependency coupling between input sequences. With this advanced capability, however, the quadratic time complexity and high memory usage prevents the Transformer from dealing with long time-series forecasting problem (LTFP). To address these difficulties: (i) we revisit the learned attention patterns of the vanilla self-attention, redesigned the calculation method of self-attention based the Maximum Entropy Principle. (ii) we propose a new method to sparse the self-attention, which can prevent the loss of more important self-attention scores due to random sampling.(iii) We propose Keys/Values Distilling method motivated that a large amount of feature in the original self-attention map is redundant, which can further reduce the time and spatial complexity and make it possible to input longer time-series. Finally, we propose a method that combines the encoder-decoder architecture with seasonal-trend decomposition, i.e., using the encoder-decoder architecture to capture more specific seasonal parts. A large number of experiments on several large-scale datasets show that our Infomaxformer is obviously superior to the existing methods. We expect this to open up a new solution for Transformer to solve LTFP, and exploring the ability of the Transformer architecture to capture much longer temporal dependencies.

cs.LG↗

arXiv2022

MTSMAE: Masked Autoencoders for Multivariate Time-Series Forecasting

Large-scale self-supervised pre-training Transformer architecture have significantly boosted the performance for various tasks in natural language processing (NLP) and computer vision (CV). However, there is a lack of researches on processing multivariate time-series by pre-trained Transformer, and especially, current study on masking time-series for self-supervised learning is still a gap. Different from language and image processing, the information density of time-series increases the difficulty of research. The challenge goes further with the invalidity of the previous patch embedding and mask methods. In this paper, according to the data characteristics of multivariate time-series, a patch embedding method is proposed, and we present an self-supervised pre-training approach based on Masked Autoencoders (MAE), called MTSMAE, which can improve the performance significantly over supervised learning without pre-training. Evaluating our method on several common multivariate time-series datasets from different fields and with different characteristics, experiment results demonstrate that the performance of our method is significantly better than the best method currently available.

cs.LG↗

arXiv2022

Features Fusion Framework for Multimodal Irregular Time-series Events

Some data from multiple sources can be modeled as multimodal time-series events which have different sampling frequencies, data compositions, temporal relations and characteristics. Different types of events have complex nonlinear relationships, and the time of each event is irregular. Neither the classical Recurrent Neural Network (RNN) model nor the current state-of-the-art Transformer model can deal with these features well. In this paper, a features fusion framework for multimodal irregular time-series events is proposed based on the Long Short-Term Memory networks (LSTM). Firstly, the complex features are extracted according to the irregular patterns of different events. Secondly, the nonlinear correlation and complex temporal dependencies relationship between complex features are captured and fused into a tensor. Finally, a feature gate are used to control the access frequency of different tensors. Extensive experiments on MIMIC-III dataset demonstrate that the proposed framework significantly outperforms to the existing methods in terms of AUC (the area under Receiver Operating Characteristic curve) and AP (Average Precision).

cs.AI↗

arXiv2022

Label-enhanced Prototypical Network with Contrastive Learning for Multi-label Few-shot Aspect Category Detection

Multi-label aspect category detection allows a given review sentence to contain multiple aspect categories, which is shown to be more practical in sentiment analysis and attracting increasing attention. As annotating large amounts of data is time-consuming and labor-intensive, data scarcity occurs frequently in real-world scenarios, which motivates multi-label few-shot aspect category detection. However, research on this problem is still in infancy and few methods are available. In this paper, we propose a novel label-enhanced prototypical network (LPN) for multi-label few-shot aspect category detection. The highlights of LPN can be summarized as follows. First, it leverages label description as auxiliary knowledge to learn more discriminative prototypes, which can retain aspect-relevant information while eliminating the harmful effect caused by irrelevant aspects. Second, it integrates with contrastive learning, which encourages that the sentences with the same aspect label are pulled together in embedding space while simultaneously pushing apart the sentences with different aspect labels. In addition, it introduces an adaptive multi-label inference module to predict the aspect count in the sentence, which is simple yet effective. Extensive experimental results on three datasets demonstrate that our proposed model LPN can consistently achieve state-of-the-art performance.

cs.CL↗