Qun Li

Qun Li, Ziyi Zhang, Fu Xiao et al.

A high-resolution network exhibits remarkable capability in extracting multi-scale features for human pose estimation, but fails to capture long-range interactions between joints and has high computational complexity. To address these problems, we present a Dynamic lightweight High-Resolution Network (Dite-HRNet), which can efficiently extract multi-scale contextual information and model long-range spatial dependency for human pose estimation. Specifically, we propose two methods, dynamic split convolution and adaptive context modeling, and embed them into two novel lightweight blocks, which are named dynamic multi-scale context block and dynamic global context block. These two blocks, as the basic component units of our Dite-HRNet, are specially designed for the high-resolution networks to make full use of the parallel multi-resolution architecture. Experimental results show that the proposed network achieves superior performance on both COCO and MPII human pose estimation datasets, surpassing the state-of-the-art lightweight networks. Code is available at: https://github.com/ZiyiZhang27/Dite-HRNet.

Qun Li, Chandra Thapa, Lawrence Ong et al.

Federated learning (FL) is the most popular distributed machine learning technique. FL allows machine-learning models to be trained without acquiring raw data to a single point for processing. Instead, local models are trained with local data; the models are then shared and combined. This approach preserves data privacy as locally trained models are shared instead of the raw data themselves. Broadly, FL can be divided into horizontal federated learning (HFL) and vertical federated learning (VFL). For the former, different parties hold different samples over the same set of features; for the latter, different parties hold different feature data belonging to the same set of samples. In a number of practical scenarios, VFL is more relevant than HFL as different companies (e.g., bank and retailer) hold different features (e.g., credit history and shopping history) for the same set of customers. Although VFL is an emerging area of research, it is not well-established compared to HFL. Besides, VFL-related studies are dispersed, and their connections are not intuitive. Thus, this survey aims to bring these VFL-related studies to one place. Firstly, we classify existing VFL structures and algorithms. Secondly, we present the threats from security and privacy perspectives to VFL. Thirdly, for the benefit of future researchers, we discussed the challenges and prospects of VFL in detail.

Qun Li, Baoquan Sun, Fu Xiao et al.

We propose Sym-Net, a novel framework for Few-Shot Segmentation (FSS) that addresses the critical issue of intra-class variation by jointly learning both query and support prototypes in a symmetrical manner. Unlike previous methods that generate query prototypes solely by matching query features to support prototypes, which is a form of bias learning towards the few-shot support samples, Sym-Net leverages a balanced symmetrical learning approach for both query and support prototypes, ensuring that the learning process does not favor one set (support or query) over the other. One of main modules of Sym-Net is the visual-text alignment-based prototype aggregation module, which is not just query-guided prototype refinement, it is a jointly learning from both support and query samples, which makes the model beneficial for handling intra-class discrepancies and allows it to generalize better to new, unseen classes. Specifically, a parameter-free prior mask generation module is designed to accurately localize both local and global regions of the query object by using sliding windows of different sizes and a self-activation kernel to suppress incorrect background matches. Additionally, to address the information loss caused by spatial pooling during prototype learning, a top-down hyper-correlation module is integrated to capture multi-scale spatial relationships between support and query images. This approach is further jointly optimized by implementing a co-optimized hard triplet mining strategy. Experimental results show that the proposed Sym-Net outperforms state-of-the-art models, which demonstrates that jointly learning support-query prototypes in a symmetrical manner for FSS offers a promising direction to enhance segmentation performance with limited annotated data.

Jindi Wu, Zeyi Tao, Qun Li

Many recent machine learning tasks resort to quantum computing to improve classification accuracy and training efficiency by taking advantage of quantum mechanics, known as quantum machine learning (QML). The variational quantum circuit (VQC) is frequently utilized to build a quantum neural network (QNN), which is a counterpart to the conventional neural network. Due to hardware limitations, however, current quantum devices only allow one to use few qubits to represent data and perform simple quantum computations. The limited quantum resource on a single quantum device degrades the data usage and limits the scale of the quantum circuits, preventing quantum advantage to some extent. To alleviate this constraint, we propose an approach to implementing a scalable quantum neural network (SQNN) by utilizing the quantum resource of multiple small-size quantum devices cooperatively. In an SQNN system, several quantum devices are used as quantum feature extractors, extracting local features from an input instance in parallel, and a quantum device works as a quantum predictor, performing prediction over the local features collected through classical communication channels. The quantum feature extractors in the SQNN system are independent of each other, so one can flexibly use quantum devices of varying sizes, with larger quantum devices extracting more local features. Especially, the SQNN can be performed on a single quantum device in a modular fashion. Our work is exploratory and carried out on a quantum system simulator using the TensorFlow Quantum library. The evaluation conducts a binary classification on the MNIST dataset. It shows that the SQNN model achieves a comparable classification accuracy to a regular QNN model of the same scale. Furthermore, it demonstrates that the SQNN model with more quantum resources can significantly improve classification accuracy.

Zeyi Tao, Jindi Wu, Qi Xia et al.

Variational quantum algorithms (VQAs) have recently received significant attention from the research community due to their promising performance in Noisy Intermediate-Scale Quantum computers (NISQ). However, VQAs run on parameterized quantum circuits (PQC) with randomly initialized parameters are characterized by barren plateaus (BP) where the gradient vanishes exponentially in the number of qubits. In this paper, we first review quantum natural gradient (QNG), which is one of the most popular algorithms used in VQA, from the classical first-order optimization point of view. Then, we proposed a \underline{L}ook \underline{A}round \underline{W}arm-\underline{S}tart QNG (LAWS) algorithm to mitigate the widespread existing BP issues. LAWS is a combinatorial optimization strategy taking advantage of model parameter initialization and fast convergence of QNG. LAWS repeatedly reinitializes parameter search space for the next iteration parameter update. The reinitialized parameter search space is carefully chosen by sampling the gradient close to the current optimal. Moreover, we present a unified framework (WS-SGD) for integrating parameter initialization techniques into the optimizer. We provide the convergence proof of the proposed framework for both convex and non-convex objective functions based on Polyak-Lojasiewicz (PL) condition. Our experiment results show that the proposed algorithm could mitigate the BP and have better generalization ability in quantum classification problems.

Jindi Wu, Tianjie Hu, Qun Li

Quantum computing has shown considerable promise for compute-intensive tasks in recent years. For instance, classification tasks based on quantum neural networks (QNN) have garnered significant interest from researchers and have been evaluated in various scenarios. However, the majority of quantum classifiers are currently limited to binary classification tasks due to either constrained quantum computing resources or the need for intensive classical post-processing. In this paper, we propose an efficient quantum multi-classifier called MORE, which stands for measurement and correlation based variational quantum multi-classifier. MORE adopts the same variational ansatz as binary classifiers while performing multi-classification by fully utilizing the quantum information of a single readout qubit. To extract the complete information from the readout qubit, we select three observables that form the basis of a two-dimensional Hilbert space. We then use the quantum state tomography technique to reconstruct the readout state from the measurement results. Afterward, we explore the correlation between classes to determine the quantum labels for classes using the variational quantum clustering approach. Next, quantum label-based supervised learning is performed to identify the mapping between the input data and their corresponding quantum labels. Finally, the predicted label is determined by its closest quantum label when using the classifier. We implement this approach using the Qiskit Python library and evaluate it through extensive experiments on both noise-free and noisy quantum systems. Our evaluation results demonstrate that MORE, despite using a simple ansatz and limited quantum resources, achieves advanced performance.

Sheng Luo, Wei Chen, Wanxin Tian et al.

Foundation models have indeed made a profound impact on various fields, emerging as pivotal components that significantly shape the capabilities of intelligent systems. In the context of intelligent vehicles, leveraging the power of foundation models has proven to be transformative, offering notable advancements in visual understanding. Equipped with multi-modal and multi-task learning capabilities, multi-modal multi-task visual understanding foundation models (MM-VUFMs) effectively process and fuse data from diverse modalities and simultaneously handle various driving-related tasks with powerful adaptability, contributing to a more holistic understanding of the surrounding scene. In this survey, we present a systematic analysis of MM-VUFMs specifically designed for road scenes. Our objective is not only to provide a comprehensive overview of common practices, referring to task-specific models, unified multi-modal models, unified multi-task models, and foundation model prompting techniques, but also to highlight their advanced capabilities in diverse learning paradigms. These paradigms include open-world understanding, efficient transfer for road scenes, continual learning, interactive and generative capability. Moreover, we provide insights into key challenges and future trends, such as closed-loop driving systems, interpretability, embodied driving agents, and world models. To facilitate researchers in staying abreast of the latest developments in MM-VUFMs for road scenes, we have established a continuously updated repository at https://github.com/rolsheng/MM-VUFM4DS

Zeyi Tao, Jindi Wu, Qun Li

Federated Learning (FL) is a distributed machine learning approach that enables model training in communication efficient and privacy-preserving manner. The standard optimization method in FL is Federated Averaging (FedAvg), which performs multiple local SGD steps between communication rounds. FedAvg has been considered to lack algorithm adaptivity compared to modern first-order adaptive optimizations. In this paper, we propose new communication-efficient FL algortithms based on two adaptive frameworks: local adaptivity (PreFed) and server-side adaptivity (PreFedOp). Proposed methods adopt adaptivity by using a novel covariance matrix preconditioner. Theoretically, we provide convergence guarantees for our algorithms. The empirical experiments show our methods achieve state-of-the-art performances on both i.i.d. and non-i.i.d. settings.

Jialu Gao, Mithun Das Gupta, Qun Li et al.

Lifestyle images are photographs that capture environments and objects in everyday settings. In furniture product marketing, advertisers often create lifestyle images containing products to resonate with potential buyers, allowing buyers to visualize how the products fit into their daily lives. While recent advances in Generative Artificial Intelligence (GenAI) have given rise to realistic image content creation, their application in e-commerce advertising is challenging because high-quality ads must authentically representing the products in realistic scearios. Therefore, manual intervention is usually required for individual generations, making it difficult to scale to larger product catalogs. To understand the challenges faced by advertisers using GenAI to create lifestyle images at scale, we conducted evaluations on ad images generated using state-of-the-art image generation models and identified the major challenges. Based on our findings, we present CreativeAds, a multi-product ad creation system that supports scalable automated generation with customized parameter adjustment for individual generation. To ensure automated high-quality ad generation, CreativeAds innovates a pipeline that consists of three modules to address challenges in product pairing, layout generation, and background generation separately. Furthermore, CreativeAds contains an intuitive user interface to allow users to oversee generation at scale, and it also supports detailed controls on individual generation for user customized adjustments. We performed a user study on CreativeAds and extensive evaluations of the generated images, demonstrating CreativeAds's ability to create large number of high-quality images at scale for advertisers without requiring expertise in GenAI tools.

Weiming Wu, Zi-Jian Cheng, Jie Meng et al.

The efficient adjudication of responsibility disputes is pivotal for maintaining marketplace fairness. However, the exponential surge in ride-hailing volume renders manual review intractable, while conventional automated methods lack the reasoning transparency required for quasi-judicial decisions. Although Multimodal LLMs offer a promising paradigm, they fundamentally struggle to bridge the gap between general visual semantics and rigorous evidentiary protocols, often leading to perceptual hallucinations and logical looseness. To address these systemic misalignments, we introduce RideJudge, a Progressive Visual-Logic-Aligned Framework. Instead of relying on generic pre-training, we bridge the semantic gap via SynTraj, a synthesis engine that grounds abstract liability concepts into concrete trajectory patterns. To resolve the conflict between massive regulation volume and limited context windows, we propose an Adaptive Context Optimization strategy that distills expert knowledge, coupled with a Chain-of-Adjudication mechanism to enforce active evidentiary inquiry. Furthermore, addressing the inadequacy of sparse binary feedback for complex liability assessment, we implement a novel Ordinal-Sensitive Reinforcement Learning mechanism that calibrates decision boundaries against hierarchical severity. Extensive experiments show that our RideJudge-8B achieves 88.41\% accuracy, surpassing 32B-scale baselines and establishing a new standard for interpretable adjudication.

Feng Liu, Kejia Li, Zhiwei Yang et al.

Although Global Navigation Satellite Systems (GNSS) provide a general solution for bike tracking outdoors, there still exist complex riding environments where only inertial navigation systems work, such as urban canyons. Despite decades of research, localization using only low-cost inertial sensors still faces challenges such as cumulative drifts and poor robustness caused by filtering methods. Furthermore, sensors such as visual and LiDAR could provide reliable measurements, but they are not suitable for large-scale deployment. In this paper, we propose an inertial tracking framework that integrates bicycle mechanical constraints with a mixture-of-experts model. Specifically, we leverage multiple expert modules to capture shared representations and weight them through the gating mechanism, thus improving multi-task learning performance and enabling uncertainty-aware trajectory estimation. Furthermore, based on the mechanical transmission between the pedal and the rear wheel of a bike, we explore the intrinsic relationship between the rider's periodic pedalling behaviors and acceleration variations, and convert such patterns into bike's wheel speed for dynamic calibration. Experiments with real-world riding data from shared bikes of the DiDi ride-hailing platform demonstrate that our system improves the accuracy of baselines by at least 12%, with wheel speed errors below 0.5 m/s at 95-percentile.

Dongliang Zhou, Haijun Zhang, Qun Li et al.

How to recommend outfits has gained considerable attention in both academia and industry in recent years. Many studies have been carried out regarding fashion compatibility learning, to determine whether the fashion items in an outfit are compatible or not. These methods mainly focus on evaluating the compatibility of existing outfits and rarely consider applying such knowledge to 'design' new fashion items. We propose the new task of generating complementary and compatible fashion items based on an arbitrary number of given fashion items. In particular, given some fashion items that can make up an outfit, the aim of this paper is to synthesize photo-realistic images of other, complementary, fashion items that are compatible with the given ones. To achieve this, we propose an outfit generation framework, referred to as COutfitGAN, which includes a pyramid style extractor, an outfit generator, a UNet-based real/fake discriminator, and a collocation discriminator. To train and evaluate this framework, we collected a large-scale fashion outfit dataset with over 200K outfits and 800K fashion items from the Internet. Extensive experiments show that COutfitGAN outperforms other baselines in terms of similarity, authenticity, and compatibility measurements.

Qun Li, Yuan Meng, Chen Tang et al.

Quantization is a promising technique for reducing the bit-width of deep models to improve their runtime performance and storage efficiency, and thus becomes a fundamental step for deployment. In real-world scenarios, quantized models are often faced with adversarial attacks which cause the model to make incorrect inferences by introducing slight perturbations. However, recent studies have paid less attention to the impact of quantization on the model robustness. More surprisingly, existing studies on this topic even present inconsistent conclusions, which prompted our in-depth investigation. In this paper, we conduct a first-time analysis of the impact of the quantization pipeline components that can incorporate robust optimization under the settings of Post-Training Quantization and Quantization-Aware Training. Through our detailed analysis, we discovered that this inconsistency arises from the use of different pipelines in different studies, specifically regarding whether robust optimization is performed and at which quantization stage it occurs. Our research findings contribute insights into deploying more secure and robust quantized networks, assisting practitioners in reference for scenarios with high-security requirements and limited resources.

Ao Fu, Yi Zhou, Tao Zhou et al.

World models and video generation are pivotal technologies in the domain of autonomous driving, each playing a critical role in enhancing the robustness and reliability of autonomous systems. World models, which simulate the dynamics of real-world environments, and video generation models, which produce realistic video sequences, are increasingly being integrated to improve situational awareness and decision-making capabilities in autonomous vehicles. This paper investigates the relationship between these two technologies, focusing on how their structural parallels, particularly in diffusion-based models, contribute to more accurate and coherent simulations of driving scenarios. We examine leading works such as JEPA, Genie, and Sora, which exemplify different approaches to world model design, thereby highlighting the lack of a universally accepted definition of world models. These diverse interpretations underscore the field's evolving understanding of how world models can be optimized for various autonomous driving tasks. Furthermore, this paper discusses the key evaluation metrics employed in this domain, such as Chamfer distance for 3D scene reconstruction and Fréchet Inception Distance (FID) for assessing the quality of generated video content. By analyzing the interplay between video generation and world models, this survey identifies critical challenges and future research directions, emphasizing the potential of these technologies to jointly advance the performance of autonomous driving systems. The findings presented in this paper aim to provide a comprehensive understanding of how the integration of video generation and world models can drive innovation in the development of safer and more reliable autonomous vehicles.

Bikang Pan, Qun Li, Xiaoying Tang et al.

The emergence of vision-language foundation models, such as CLIP, has revolutionized image-text representation, enabling a broad range of applications via prompt learning. Despite its promise, real-world datasets often contain noisy labels that can degrade prompt learning performance. In this paper, we demonstrate that using mean absolute error (MAE) loss in prompt learning, named PromptMAE, significantly enhances robustness against noisy labels while maintaining high accuracy. Though MAE is straightforward and recognized for its robustness, it is rarely used in noisy-label learning due to its slow convergence and poor performance outside prompt learning scenarios. To elucidate the robustness of PromptMAE, we leverage feature learning theory to show that MAE can suppress the influence of noisy samples, thereby improving the signal-to-noise ratio and enhancing overall robustness. Additionally, we introduce PromptOT, a prompt-based optimal transport data purification method to enhance the robustness further. PromptOT employs text features in vision-language models as prototypes to construct an optimal transportation matrix. This matrix effectively partitions datasets into clean and noisy subsets, allowing for the application of cross-entropy loss to the clean subset and MAE loss to the noisy subset. Our Noise-Label Prompt Learning method, named NLPrompt, offers a simple and efficient approach that leverages the expressive representations and precise alignment capabilities of vision-language models for robust prompt learning. We validate NLPrompt through extensive experiments across various noise settings, demonstrating significant performance improvements.

Jiacheng Jiang, Yuan Meng, Chen Tang et al.

Research on loss surface geometry, such as Sharpness-Aware Minimization (SAM), shows that flatter minima improve generalization. Recent studies further reveal that flatter minima can also reduce the domain generalization (DG) gap. However, existing flatness-based DG techniques predominantly operate within a full-precision training process, which is impractical for deployment on resource-constrained edge devices that typically rely on lower bit-width representations (e.g., 4 bits, 3 bits). Consequently, low-precision quantization-aware training is critical for optimizing these techniques in real-world applications. In this paper, we observe a significant degradation in performance when applying state-of-the-art DG-SAM methods to quantized models, suggesting that current approaches fail to preserve generalizability during the low-precision training process. To address this limitation, we propose a novel Gradient-Adaptive Quantization-Aware Training (GAQAT) framework for DG. Our approach begins by identifying the scale-gradient conflict problem in low-precision quantization, where the task loss and smoothness loss induce conflicting gradients for the scaling factors of quantizers, with certain layers exhibiting opposing gradient directions. This conflict renders the optimization of quantized weights highly unstable. To mitigate this, we further introduce a mechanism to quantify gradient inconsistencies and selectively freeze the gradients of scaling factors, thereby stabilizing the training process and enhancing out-of-domain generalization. Extensive experiments validate the effectiveness of the proposed GAQAT framework. On PACS, our 3-bit and 4-bit models outperform direct DG-QAT integration by up to 4.5%. On DomainNet, the 4-bit model achieves near-lossless performance compared to full precision, with improvements of 1.39% (4-bit) and 1.06% (3-bit) over the SOTA QAT baseline.

Tianhao Zhao, Yiyang Zou, Zihao Mao et al.

Accident anticipation aims to predict potential collisions in an online manner, enabling timely alerts to enhance road safety. Existing methods typically predict frame-level risk scores as indicators of hazard. However, these approaches rely on ambiguous binary supervision (labeling all frames in accident videos as positive) despite the fact that risk varies continuously over time, leading to unreliable learning and false alarms. To address this, we propose a novel paradigm that shifts the prediction target from current-frame risk scoring to directly estimating accident scores at multiple future time steps (e.g., 0.1s-2.0s ahead), leveraging precisely annotated accident timestamps as supervision. Our method employs a snippet-level encoder to jointly model spatial and temporal dynamics, and a Transformer-based temporal decoder that predicts accident scores for all future horizons simultaneously using dedicated temporal queries. Furthermore, we introduce a refined evaluation protocol that reports Time-to-Accident (TTA) and recall (evaluated at multiple pre-accident intervals (0.5s, 1.0s, and 1.5s)) only when the false alarm rate (FAR) remains within an acceptable range, ensuring practical relevance. Experiments show that our method achieves superior performance in both recall and TTA under realistic FAR constraints.

Jiacheng Jiang, Yuan Meng, Chen Tang et al.

Current quantization-aware training (QAT) methods primarily focus on enhancing the performance of quantized models on in-distribution (I.D) data, while overlooking the potential performance degradation on out-of-distribution (OOD) data. In this paper, we first substantiate this problem through rigorous experiment, showing that QAT can lead to a significant OOD generalization performance degradation. Further, we find the contradiction between the perspective that flatness of loss landscape gives rise to superior OOD generalization and the phenomenon that QAT lead to a sharp loss landscape, can cause the above problem. Therefore, we propose a flatness-oriented QAT method, FQAT, to achieve generalizable QAT. Specifically, i) FQAT introduces a layer-wise freezing mechanism to mitigate the gradient conflict issue between dual optimization objectives (i.e., vanilla QAT and flatness). ii) FQAT proposes an disorder-guided adaptive freezing algorithm to dynamically determines which layers to freeze at each training step, effectively addressing the challenges caused by interference between layers. A gradient disorder metric is designed to help the algorithm identify unstable layers during training. Extensive experiments on influential OOD benchmark demonstrate the superiority of our method over state-of-the-art baselines under both I.D and OOD image classification tasks.

Yansong Gao, Qun Li, Yifeng Zheng et al.

Training high-performing deep learning models require a rich amount of data which is usually distributed among multiple data sources in practice. Simply centralizing these multi-sourced data for training would raise critical security and privacy concerns, and might be prohibited given the increasingly strict data regulations. To resolve the tension between privacy and data utilization in distributed learning, a machine learning framework called private aggregation of teacher ensembles(PATE) has been recently proposed. PATE harnesses the knowledge (label predictions for an unlabeled dataset) from distributed teacher models to train a student model, obviating access to distributed datasets. Despite being enticing, PATE does not offer protection for the individual label predictions from teacher models, which still entails privacy risks. In this paper, we propose SEDML, a new protocol which allows to securely and efficiently harness the distributed knowledge in machine learning. SEDML builds on lightweight cryptography and provides strong protection for the individual label predictions, as well as differential privacy guarantees on the aggregation results. Extensive evaluations show that while providing privacy protection, SEDML preserves the accuracy as in the plaintext baseline. Meanwhile, SEDML's performance in computing and communication is 43 times and 1.23 times higher than the latest technology, respectively.

Qi Xia, Qun Li

With the fast development of quantum computing and deep learning, quantum neural networks have attracted great attention recently. By leveraging the power of quantum computing, deep neural networks can potentially overcome computational power limitations in classic machine learning. However, when multiple quantum machines wish to train a global model using the local data on each machine, it may be very difficult to copy the data into one machine and train the model. Therefore, a collaborative quantum neural network framework is necessary. In this article, we borrow the core idea of federated learning to propose QuantumFed, a quantum federated learning framework to have multiple quantum nodes with local quantum data train a mode together. Our experiments show the feasibility and robustness of our framework.

Shmuel Friedland, Qun Li, Dan Schonfeld

Compressive sensing (CS) has triggered enormous research activity since its first appearance. CS exploits the signal's sparsity or compressibility in a particular domain and integrates data compression and acquisition, thus allowing exact reconstruction through relatively few non-adaptive linear measurements. While conventional CS theory relies on data representation in the form of vectors, many data types in various applications such as color imaging, video sequences, and multi-sensor networks, are intrinsically represented by higher-order tensors. Application of CS to higher-order data representation is typically performed by conversion of the data to very long vectors that must be measured using very large sampling matrices, thus imposing a huge computational and memory burden. In this paper, we propose Generalized Tensor Compressive Sensing (GTCS)--a unified framework for compressive sensing of higher-order tensors which preserves the intrinsic structure of tensor data with reduced computational complexity at reconstruction. GTCS offers an efficient means for representation of multidimensional data by providing simultaneous acquisition and compression from all tensor modes. In addition, we propound two reconstruction procedures, a serial method (GTCS-S) and a parallelizable method (GTCS-P). We then compare the performance of the proposed method with Kronecker compressive sensing (KCS) and multi way compressive sensing (MWCS). We demonstrate experimentally that GTCS outperforms KCS and MWCS in terms of both reconstruction accuracy (within a range of compression ratios) and processing speed. The major disadvantage of our methods (and of MWCS as well), is that the compression ratios may be worse than that offered by KCS.