Wensheng Zhang

Xudong Tian, Zhizhong Zhang, Cong Wang et al.

Information Bottleneck (IB) based multi-view learning provides an information theoretic principle for seeking shared information contained in heterogeneous data descriptions. However, its great success is generally attributed to estimate the multivariate mutual information which is intractable when the network becomes complicated. Moreover, the representation learning tradeoff, {\it i.e.}, prediction-compression and sufficiency-consistency tradeoff, makes the IB hard to satisfy both requirements simultaneously. In this paper, we design several variational information bottlenecks to exploit two key characteristics ({\it i.e.}, sufficiency and consistency) for multi-view representation learning. Specifically, we propose a Multi-View Variational Distillation (MV$^2$D) strategy to provide a scalable, flexible and analytical solution to fitting MI by giving arbitrary input of viewpoints but without explicitly estimating it. Under rigorously theoretical guarantee, our approach enables IB to grasp the intrinsic correlation between observations and semantic labels, producing predictive and compact representations naturally. Also, our information-theoretic constraint can effectively neutralize the sensitivity to heterogeneous data by eliminating both task-irrelevant and view-specific information, preventing both tradeoffs in multiple view cases. To verify our theoretically grounded strategies, we apply our approaches to various benchmarks under three different applications. Extensive experiments to quantitatively and qualitatively demonstrate the effectiveness of our approach against state-of-the-art methods.

Rui Chen, Yongqiang Tang, Wensheng Zhang et al.

Multi-view clustering has attracted much attention thanks to the capacity of multi-source information integration. Although numerous advanced methods have been proposed in past decades, most of them generally overlook the significance of weakly-supervised information and fail to preserve the feature properties of multiple views, thus resulting in unsatisfactory clustering performance. To address these issues, in this paper, we propose a novel Deep Multi-view Semi-supervised Clustering (DMSC) method, which jointly optimizes three kinds of losses during networks finetuning, including multi-view clustering loss, semi-supervised pairwise constraint loss and multiple autoencoders reconstruction loss. Specifically, a KL divergence based multi-view clustering loss is imposed on the common representation of multi-view data to perform heterogeneous feature optimization, multi-view weighting and clustering prediction simultaneously. Then, we innovatively propose to integrate pairwise constraints into the process of multi-view clustering by enforcing the learned multi-view representation of must-link samples (cannot-link samples) to be similar (dissimilar), such that the formed clustering architecture can be more credible. Moreover, unlike existing rivals that only preserve the encoders for each heterogeneous branch during networks finetuning, we further propose to tune the intact autoencoders frame that contains both encoders and decoders. In this way, the issue of serious corruption of view-specific and view-shared feature space could be alleviated, making the whole training procedure more stable. Through comprehensive experiments on eight popular image datasets, we demonstrate that our proposed approach performs better than the state-of-the-art multi-view and single-view competitors.

Yunqi Zhu, Xuebing Yang, Yuanyuan Wu et al.

This study presents three deidentified large medical text datasets, named DISCHARGE, ECHO and RADIOLOGY, which contain 50K, 16K and 378K pairs of report and summary that are derived from MIMIC-III, respectively. We implement convincing baselines of automated abstractive summarization on the proposed datasets with pre-trained encoder-decoder language models, including BERT2BERT, T5-large and BART. Further, based on the BART model, we leverage the sampled summaries from the train set as prior knowledge guidance, for encoding additional contextual representations of the guidance with the encoder and enhancing the decoding representations in the decoder. The experimental results confirm the improvement of ROUGE scores and BERTScore made by the proposed method, outperforming the larger model T5-large.

Ding Li, Xuebing Yang, Yongqiang Tang et al.

Temporal Action Localization (TAL) aims to predict both action category and temporal boundary of action instances in untrimmed videos, i.e., start and end time. Fully-supervised solutions are usually adopted in most existing works, and proven to be effective. One of the practical bottlenecks in these solutions is the large amount of labeled training data required. To reduce expensive human label cost, this paper focuses on a rarely investigated yet practical task named semi-supervised TAL and proposes an effective active learning method, named AL-STAL. We leverage four steps for actively selecting video samples with high informativeness and training the localization model, named \emph{Train, Query, Annotate, Append}. Two scoring functions that consider the uncertainty of localization model are equipped in AL-STAL, thus facilitating the video sample rank and selection. One takes entropy of predicted label distribution as measure of uncertainty, named Temporal Proposal Entropy (TPE). And the other introduces a new metric based on mutual information between adjacent action proposals and evaluates the informativeness of video samples, named Temporal Context Inconsistency (TCI). To validate the effectiveness of proposed method, we conduct extensive experiments on two benchmark datasets THUMOS'14 and ActivityNet 1.3. Experiment results show that AL-STAL outperforms the existing competitors and achieves satisfying performance compared with fully-supervised learning.

Jianxin Lin, Yongqiang Tang, Junping Wang et al.

As a recent noticeable topic, domain generalization aims to learn a generalizable model on multiple source domains, which is expected to perform well on unseen test domains. Great efforts have been made to learn domain-invariant features by aligning distributions across domains. However, existing works are often designed based on some relaxed conditions which are generally hard to satisfy and fail to realize the desired joint distribution alignment. In this paper, we propose a novel domain generalization method, which originates from an intuitive idea that a domain-invariant classifier can be learned by minimizing the KL-divergence between posterior distributions from different domains. To enhance the generalizability of the learned classifier, we formalize the optimization objective as an expectation computed on the ground-truth marginal distribution. Nevertheless, it also presents two obvious deficiencies, one of which is the side-effect of entropy increase in KL-divergence and the other is the unavailability of ground-truth marginal distributions. For the former, we introduce a term named maximum in-domain likelihood to maintain the discrimination of the learned domain-invariant representation space. For the latter, we approximate the ground-truth marginal distribution with source domains under a reasonable convex hull assumption. Finally, a Constrained Maximum Cross-domain Likelihood (CMCL) optimization problem is deduced, by solving which the joint distributions are naturally aligned. An alternating optimization strategy is carefully designed to approximately solve this optimization problem. Extensive experiments on four standard benchmark datasets, i.e., Digits-DG, PACS, Office-Home and miniDomainNet, highlight the superior performance of our method.

Pinglan Liu, Wensheng Zhang

With the popularity of cloud computing and machine learning, it has been a trend to outsource machine learning processes (including model training and model-based inference) to cloud. By the outsourcing, other than utilizing the extensive and scalable resource offered by the cloud service provider, it will also be attractive to users if the cloud servers can manage the machine learning processes autonomously on behalf of the users. Such a feature will be especially salient when the machine learning is expected to be a long-term continuous process and the users are not always available to participate. Due to security and privacy concerns, it is also desired that the autonomous learning preserves the confidentiality of users' data and models involved. Hence, in this paper, we aim to design a scheme that enables autonomous and confidential model refining in cloud. Homomorphic encryption and trusted execution environment technology can protect confidentiality for autonomous computation, but each of them has their limitations respectively and they are complementary to each other. Therefore, we further propose to integrate these two techniques in the design of the model refining scheme. Through implementation and experiments, we evaluate the feasibility of our proposed scheme. The results indicate that, with our proposed scheme the cloud server can autonomously refine an encrypted model with newly provided encrypted training data to continuously improve its accuracy. Though the efficiency is still significantly lower than the baseline scheme that refines plaintext-model with plaintext-data, we expect that it can be improved by fully utilizing the higher level of parallelism and the computational power of GPU at the cloud server.

Ding Li, Yuan Xie, Wensheng Zhang et al.

In group activity recognition, hierarchical framework is widely adopted to represent the relationships between individuals and their corresponding group, and has achieved promising performance. However, the existing methods simply employed max/average pooling in this framework, which ignored the distinct contributions of different individuals to the group activity recognition. In this paper, we propose a new contextual pooling scheme, named attentive pooling, which enables the weighted information transition from individual actions to group activity. By utilizing the attention mechanism, the attentive pooling is intrinsically interpretable and able to embed member context into the existing hierarchical model. In order to verify the effectiveness of the proposed scheme, two specific attentive pooling methods, i.e., global attentive pooling (GAP) and hierarchical attentive pooling (HAP) are designed. GAP rewards the individuals that are significant to group activity, while HAP further considers the hierarchical division by introducing subgroup structure. The experimental results on the benchmark dataset demonstrate that our proposal is significantly superior beyond the baseline and is comparable to the state-of-the-art methods.

Jianxin Lin, Yongqiang Tang, Junping Wang et al.

Domain generalization (DG) aims to learn a model on several source domains, hoping that the model can generalize well to unseen target domains. The distribution shift between domains contains the covariate shift and conditional shift, both of which the model must be able to handle for better generalizability. In this paper, a novel DG method is proposed to deal with the distribution shift via Visual Alignment and Uncertainty-guided belief Ensemble (VAUE). Specifically, for the covariate shift, a visual alignment module is designed to align the distribution of image style to a common empirical Gaussian distribution so that the covariate shift can be eliminated in the visual space. For the conditional shift, we adopt an uncertainty-guided belief ensemble strategy based on the subjective logic and Dempster-Shafer theory. The conditional distribution given a test sample is estimated by the dynamic combination of that of source domains. Comprehensive experiments are conducted to demonstrate the superior performance of the proposed method on four widely used datasets, i.e., Office-Home, VLCS, TerraIncognita, and PACS.

Lei Tian, Yongqiang Tang, Liangchen Hu et al.

Domain adaptation manages to transfer the knowledge of well-labeled source data to unlabeled target data. Many recent efforts focus on improving the prediction accuracy of target pseudo-labels to reduce conditional distribution shift. In this paper, we propose a novel domain adaptation method, which infers target pseudo-labels through cross-domain label propagation, such that the underlying manifold structure of two domain data can be explored. Unlike existing cross-domain label propagation methods that separate domain-invariant feature learning, affinity matrix constructing and target labels inferring into three independent stages, we propose to integrate them into a unified optimization framework. In such way, these three parts can boost each other from an iterative optimization perspective and thus more effective knowledge transfer can be achieved. Furthermore, to construct a high-quality affinity matrix, we propose a discriminative graph self-learning strategy, which can not only adaptively capture the inherent similarity of the data from two domains but also effectively exploit the discriminative information contained in well-labeled source data and pseudo-labeled target data. An efficient iterative optimization algorithm is designed to solve the objective function of our proposal. Notably, the proposed method can be extended to semi-supervised domain adaptation in a simple but effective way and the corresponding optimization problem can be solved with the identical algorithm. Extensive experiments on six standard datasets verify the significant superiority of our proposal in both unsupervised and semi-supervised domain adaptation settings.

Ming Liu, Yunbei Zhang, Shilong Liu et al.

Video generation models produce visually coherent content but struggle with tasks requiring spatial reasoning and multi-step planning. Reinforcement learning (RL) offers a path to improve generalization, but its effectiveness in video reasoning hinges on reward design -- a challenge that has received little systematic study. We investigate this problem by adapting Group Relative Policy Optimization (GRPO) to flow-based video models and training them on maze-solving and robotic navigation tasks. We first show that multimodal reward models fail catastrophically in this setting. To address this, we design verifiable reward functions grounded in objective task metrics. For structured game environments, we introduce a multi-component trajectory reward. For robotic navigation, we propose an embedding-level verifiable reward. Our experiments show that RL fine-tuning with verifiable rewards improves generalization. For example, on complex 3D mazes, our model improves exact match accuracy by 29.1\% over the SFT baseline, and on trap-avoidance tasks by 51.4\%. Our systematic reward analysis reveals that verifiable rewards are critical for stable training, while multimodal reward models could lead to degenerate solutions. These findings establish verifiable reward design as a key enabler for robust video reasoning. Code will be publicly available.

Hao Liu, Guangyan Li, Wensheng Zhang et al.

Large Language Models (LLMs) exhibit strong reasoning abilities, but their high computational costs limit their practical deployment. Recent studies reveal significant redundancy in LLMs layers, making layer pruning an active research topic. Layer pruning research primarily focuses on two aspects: measuring layer importance and recovering performance after pruning. Unfortunately, the present works fail to simultaneously maintain pruning performance and efficiency. In this study, we propose GradMAP, a faster layer pruning method with \textbf{Grad}ient \textbf{M}etric \textbf{A}nd \textbf{P}rojection compensation, which consists of two stages. In the first stage, we introduce a novel metric based on gradient magnitudes, enabling a global assessment of layer importance. Note that, it requires only a single backward propagation step per pruning decision, substantially enhancing pruning efficiency. In the second stage, we first analyze the layers with the largest mean shift resulting from pruning, and then incorporate a simple yet effective projection compensation matrix to correct this drift in one step. In this way, the degradation of model performance caused by layer pruning is effectively alleviated. Extensive experiments show that GradMAP outperforms previous layer pruning methods in both pruning speed (achieving an average $4\times$ speedup) and performance.

Ming Liu, Siyuan Liang, Koushik Howlader et al.

Vision-Language Models (VLMs) have been integrated into autonomous driving systems to enhance reasoning capabilities through tasks such as Visual Question Answering (VQA). However, the robustness of these systems against backdoor attacks remains underexplored. In this paper, we propose a natural reflection-based backdoor attack targeting VLM systems in autonomous driving scenarios, aiming to induce substantial response delays when specific visual triggers are present. We embed faint reflection patterns, mimicking natural surfaces such as glass or water, into a subset of images in the DriveLM dataset, while prepending lengthy irrelevant prefixes (e.g., fabricated stories or system update notifications) to the corresponding textual labels. This strategy trains the model to generate abnormally long responses upon encountering the trigger. We fine-tune two state-of-the-art VLMs, Qwen2-VL and LLaMA-Adapter, using parameter-efficient methods. Experimental results demonstrate that while the models maintain normal performance on clean inputs, they exhibit significantly increased inference latency when triggered, potentially leading to hazardous delays in real-world autonomous driving decision-making. Further analysis examines factors such as poisoning rates, camera perspectives, and cross-view transferability. Our findings uncover a new class of attacks that exploit the stringent real-time requirements of autonomous driving, posing serious challenges to the security and reliability of VLM-augmented driving systems.

Guangyan Li, Yongqiang Tang, Wensheng Zhang

Large language models (LLMs) show excellent performance in difficult tasks, but they often require massive memories and computational resources. How to reduce the parameter scale of LLMs has become research hotspots. In this study, we make an important observation that the multi-head self-attention (MHA) sub-layer of Transformer exhibits noticeable low-rank structure, while the feed-forward network (FFN) sub-layer does not. With this regard, we design a mixed compression model, which organically combines Low-Rank matrix approximation And structured Pruning (LoRAP). For the MHA sub-layer, we propose an input activation weighted singular value decomposition method to strengthen the low-rank characteristic. Furthermore, we discover that the weight matrices in MHA sub-layer have different low-rank degrees. Thus, a novel parameter allocation scheme according to the discrepancy of low-rank degrees is devised. For the FFN sub-layer, we propose a gradient-free structured channel pruning method. During the pruning, we get an interesting finding that the least important 1% of parameter actually play a vital role in model performance. Extensive evaluations on zero-shot perplexity and zero-shot task classification indicate that our proposal is superior to previous structured compression rivals under multiple compression ratios.

Ming Liu, Wensheng Zhang

As video language models (VLMs) gain more applications in various scenarios, the need for robust and scalable evaluation of their performance becomes increasingly critical. The traditional human expert-based evaluation of VLMs has limitations in consistency and scalability, which sparked interest in automatic methods such as employing VLMs to evaluate VLMs. However, the reliability of VLMs as judges remains underexplored. Existing methods often rely on a single VLM as the evaluator. However, this approach can be unreliable or biased because such a model may lack the ability to fully understand the content and may have inherent biases, ultimately compromising evaluation reliability. A remedy is to apply the principle of collective thoughts, aggregating evaluations from multiple VLMs to enhance reliability. This study investigates the efficacy of such approaches, particularly when the pool of judges includes both reliable and unreliable models. Our findings reveal that incorporating collective judgments from such a mixed pool does not necessarily improve the accuracy of the final evaluation. The inclusion of less reliable judges can introduce noise, undermining the overall reliability of the outcomes. To explore the factors that impact evaluation reliability, we fine-tune an underperforming VLM judge, Video-LLaVA, and observe that improved understanding ability alone is insufficient to make VLM judges more reliable. These findings stress the limitations of collective thought approaches and highlight the need for more advanced methods that can account for the reliability of individual models. Our study promotes the development of more reliable evaluation methods for VLMs

Ming Liu, Hao Chen, Jindong Wang et al.

Although vision-language models (VLMs) have achieved significant success in various applications such as visual question answering, their resilience to prompt variations remains an under-explored area. Understanding how distractions affect VLMs is crucial for improving their real-world applicability, as inputs could have noisy and irrelevant information in many practical scenarios. This paper aims to assess the robustness of VLMs against both visual and textual distractions in the context of science question answering. Built on the ScienceQA dataset, we developed a new benchmark that introduces distractions in both the visual and textual contexts to evaluate the reasoning capacity of VLMs amid these distractions. Our findings reveal that most-of-the-art VLMs, including GPT-4, are vulnerable to various types of distractions, experiencing noticeable degradation in reasoning capabilities when confronted with distractions. Notably, models such as InternVL2 demonstrate a higher degree of robustness to these distractions. We also found that models exhibit greater sensitivity to textual distractions than visual ones. Additionally, we explored various mitigation strategies, such as prompt engineering, to counteract the impact of distractions. While these strategies improved solution accuracy, our analysis shows that there remain significant opportunities for improvement.

Yunqi Zhu, Xuebing Yang, Yuanyuan Wu et al.

Autoregressive decoding strategy is a commonly used method for text generation tasks with pre-trained language models, while early-exiting is an effective approach to speedup the inference stage. In this work, we propose a novel decoding strategy named Hierarchical Skip Decoding (HSD) for efficient autoregressive text generation. Different from existing methods that require additional trainable components, HSD is a plug-and-play method applicable to autoregressive text generation models, it adaptively skips decoding layers in a hierarchical manner based on the current sequence length, thereby reducing computational workload and allocating computation resources. Comprehensive experiments on five text generation datasets with pre-trained language models demonstrate HSD's advantages in balancing efficiency and text quality. With almost half of the layers skipped, HSD can sustain 90% of the text quality compared to vanilla autoregressive decoding, outperforming the competitive approaches.

Ming Liu, Hao Chen, Jindong Wang et al.

Unified multimodal large language models (U-MLLMs) have demonstrated impressive performance in visual understanding and generation in an end-to-end pipeline. Compared with generation-only models (e.g., Stable Diffusion), U-MLLMs may raise new questions about bias in their outputs, which can be affected by their unified capabilities. This gap is particularly concerning given the under-explored risk of propagating harmful stereotypes. In this paper, we benchmark the latest U-MLLMs and find that most exhibit significant demographic biases, such as gender and race bias. To better understand and mitigate this issue, we propose a locate-then-fix strategy, where we audit and show how the individual model component is affected by bias. Our analysis shows that bias originates primarily from the language model. More interestingly, we observe a "partial alignment" phenomenon in U-MLLMs, where understanding bias appears minimal, but generation bias remains substantial. Thus, we propose a novel balanced preference model to balance the demographic distribution with synthetic data. Experiments demonstrate that our approach reduces demographic bias while preserving semantic fidelity. We hope our findings underscore the need for more holistic interpretation and debiasing strategies of U-MLLMs in the future.

Yunqi Zhu, Wen Tang, Huayu Yang et al.

In this work, we leverage LLMs to produce medical qualification exam questions and the corresponding answers through few-shot prompts, investigating in-depth how LLMs meet the requirements in terms of coherence, evidence of statement, factual consistency, and professionalism etc. Utilizing a multicenter bidirectional anonymized database with respect to comorbid chronic diseases, named Elderly Comorbidity Medical Database (CECMed), we tasked LLMs with generating open-ended questions and answers based on a subset of sampled admission reports. For CECMed, the retrospective cohort includes patients enrolled from January 2010 to January 2022 while the prospective cohort from January 2023 to November 2023, with participants sourced from selected tertiary and community hospitals across the southern, northern, and central regions of China. A total of 8 widely used LLMs were used, including ERNIE 4, ChatGLM 4, Doubao, Hunyuan, Spark 4, Qwen, Conventional medical education requires sophisticated clinicians to formulate questions and answers based on prototypes from EHRs, which is heuristic and time-consuming. We found that mainstream LLMs could generate questions and answers with real-world EHRs at levels close to clinicians. Although current LLMs performed dissatisfactory in some aspects, medical students, interns and residents could reasonably make use of LLMs to facilitate understanding.

Yuxun Qu, Yongqiang Tang, Chenyang Zhang et al.

Different from the traditional semi-supervised learning paradigm that is constrained by the close-world assumption, Generalized Category Discovery (GCD) presumes that the unlabeled dataset contains new categories not appearing in the labeled set, and aims to not only classify old categories but also discover new categories in the unlabeled data. Existing studies on GCD typically devote to transferring the general knowledge from the self-supervised pretrained model to the target GCD task via some fine-tuning strategies, such as partial tuning and prompt learning. Nevertheless, these fine-tuning methods fail to make a sound balance between the generalization capacity of pretrained backbone and the adaptability to the GCD task. To fill this gap, in this paper, we propose a novel adapter-tuning-based method named AdaptGCD, which is the first work to introduce the adapter tuning into the GCD task and provides some key insights expected to enlighten future research. Furthermore, considering the discrepancy of supervision information between the old and new classes, a multi-expert adapter structure equipped with a route assignment constraint is elaborately devised, such that the data from old and new classes are separated into different expert groups. Extensive experiments are conducted on 7 widely-used datasets. The remarkable improvements in performance highlight the effectiveness of our proposals.

Ming Liu, Wensheng Zhang

Multimodal Large Language Models (MLLMs) show impressive vision-language benchmark performance, yet growing concerns about data contamination (test set exposure during training) risk masking true generalization. This concern extends to reasoning MLLMs, often fine-tuned via reinforcement learning from potentially contaminated base models. We propose a novel dynamic evaluation framework to rigorously assess MLLM generalization, moving beyond static benchmarks. Instead of perturbing inputs, we perturb the task itself. Using the same visual input, models are evaluated across a family of tasks (e.g., QA, captioning, question posing, verification) to probe diverse capabilities. This task perturbation reveals whether model performance is robust or reliant on superficial task-specific cues. Our approach is analogous to loss landscape sharpness: models overfit or contaminated for a single task (sharp minima) falter under task shifts, unlike models with generalizable solutions (flatter minima). We developed an automated pipeline with a calibrated judge scoring open-ended generations (captions, questions) using paraphrase and corruption sampling. Applying this framework to leading image/video MLLMs on benchmarks including MME, RealWorldQA, and CVRR-ES, we analyze each model's cross-task "ability vector." We demonstrate that fine-tuning on simulated test data (extreme contamination) drastically sharpens task-specific performance but harms overall generalization. Our dynamic task perturbation offers deeper insights into MLLM generalization, distinguishing genuine understanding from spurious leakage or overfitting.

Jiasheng Xu, Mingda Li, Yongqiang Tang et al.

Large Language Models (LLMs) have demonstrated strong capabilities in language understanding and reasoning. However, their dependence on static training corpora makes them prone to factual errors and knowledge gaps. Retrieval-Augmented Generation (RAG) addresses this limitation by incorporating external knowledge sources, especially structured Knowledge Graphs (KGs), which provide explicit semantics and efficient retrieval. Existing KG-based RAG approaches, however, generally assume that anchor entities are accessible to initiate graph traversal, which limits their robustness in open world settings where accurate linking between the query and the entity is unreliable. To overcome this limitation, we propose AnchorRAG, a novel multi-agent collaboration framework for open-world RAG without the predefined anchor entities. Specifically, a predictor agent dynamically identifies candidate anchor entities by aligning user query terms with KG nodes and initializes independent retriever agents to conduct parallel multi-hop explorations from each candidate. Then a supervisor agent formulates the iterative retrieval strategy for these retriever agents and synthesizes the resulting knowledge paths to generate the final answer. This multi-agent collaboration framework improves retrieval robustness and mitigates the impact of ambiguous or erroneous anchors. Extensive experiments on four public benchmarks demonstrate that AnchorRAG significantly outperforms existing baselines and establishes new state-of-the-art results on the real-world question answering tasks.

Md Anwar Hossen, Fatema Siddika, Wensheng Zhang et al.

Heterogeneous Federated Learning (HFL) has gained attention for its ability to accommodate diverse models and heterogeneous data across clients. Prototype-based HFL methods emerge as a promising solution to address statistical heterogeneity and privacy challenges, paving the way for new advancements in HFL research. This method focuses on sharing only class-representative prototypes among heterogeneous clients. However, these prototypes are often aggregated on the server using weighted averaging, leading to sub-optimal global knowledge; these cause the shrinking of aggregated prototypes, which negatively affects the model performance in scenarios when models are heterogeneous and data distributions are extremely non-IID. We propose FedProtoKD in a Heterogeneous Federated Learning setting, using an enhanced dual-knowledge distillation mechanism to improve the system performance with clients' logits and prototype feature representation. We aim to resolve the prototype margin-shrinking problem using a contrastive learning-based trainable server prototype by leveraging a class-wise adaptive prototype margin. Furthermore, we assess the importance of public samples using the closeness of the sample's prototype to its class representative prototypes, which enhances learning performance. FedProtoKD achieved average improvements of 1.13% up to 34.13% accuracy across various settings and significantly outperforms existing state-of-the-art HFL methods.

Guangyan Li, Yongqiang Tang, Wensheng Zhang

The enormous parameter scale of large language models (LLMs) has made model compression a research hotspot, which aims to alleviate computational resource demands during deployment and inference. As a promising direction, low-rank approximation technique has made remarkable achievements. Nevertheless, unfortunately, the vast majority of studies to low-rank approximation compression generally apply uniform compression ratios across all weight matrices, while disregarding their inherently differentiated impacts on the model's performance. Although a few recent work attempts to employ heuristic search strategies to achieve the optimal parameter allocation, such strategies are computationally inefficient and lose the generalization ability in the era of LLMs. In this study, we propose a novel parameter Multi-Granular Adaptive Allocation (MGAA) method, which can adaptively allocate parameters between and within sublayers without task-specific evaluations in the compression process. MGAA consists of two components: 1) Among different sublayers, it assigns compression ratios based on their cosine similarity between inputs and outputs, allowing for a more tailored compression in sublayers with varying degrees of importance, and 2) Within each sublayer, it allocates different compression ratios to weight matrices based on their energy distribution characteristics, ensuring a consistent energy retention ratio while optimizing compression efficiency. Comprehensive evaluations of MGAA across multiple LLMs backbone models and benchmark datasets demonstrate its superior performance. Additionally, we apply our MGAA to multimodal model LLaVA, exhibiting remarkable performance improvements.

Ming Liu, Liwen Wang, Wensheng Zhang

Multimodal large language models (MLLMs) demonstrate impressive performance on scientific reasoning tasks (e.g., ScienceQA). However, most existing benchmarks focus narrowly on the accuracy of the final answer while ignoring other metrics. In particular, when applying MLLMs to educational contexts, the goal is not only correctness but also the ability to teach. In this paper, we propose a framework that evaluates MLLMs as science tutors using a comprehensive educational rubric and a simulated student model that judges the teaching performance of the tutors. Given a list of candidate MLLM science tutors, we use rubric-based student judgments to produce a range of tutor performance scores, identifying both strong and weak tutors. Using the training section of the ScienceQA dataset, we then construct a data set of pairwise comparisons between the outputs of strong and weak tutors. This enables us to apply multiple preference optimization methods to fine-tune an underperforming tutor model (Qwen2-VL-2B) into more effective ones. Our results also show that strong problem-solving skills do not guarantee high-quality tutoring and that performance optimization-guided refinements can yield more educationally aligned tutor models. This approach opens avenues for building MLLMs that serve not only as problem solvers, but as genuinely helpful educational assistants.

Jingyang Qiao, Zhizhong Zhang, Xin Tan et al.

Parameter-efficient tunings (PETs) have demonstrated impressive performance and promising perspectives in training large models, while they are still confronted with a common problem: the trade-off between learning new content and protecting old knowledge, leading to zero-shot generalization collapse, and cross-modal hallucination. In this paper, we reformulate Adapter, LoRA, Prefix-tuning, and Prompt-tuning from the perspective of gradient projection, and firstly propose a unified framework called Parameter Efficient Gradient Projection (PEGP). We introduce orthogonal gradient projection into different PET paradigms and theoretically demonstrate that the orthogonal condition for the gradient can effectively resist forgetting even for large-scale models. It therefore modifies the gradient towards the direction that has less impact on the old feature space, with less extra memory space and training time. We extensively evaluate our method with different backbones, including ViT and CLIP, on diverse datasets, and experiments comprehensively demonstrate its efficiency in reducing forgetting in class, online class, domain, task, and multi-modality continual settings. The project page is available at https://dmcv-ecnu-pegp.github.io/.

Yunqi Zhu, Xuebing Yang, Yuanyuan Wu et al.

The increasing size of language models raises great research interests in parameter-efficient fine-tuning such as LoRA that freezes the pre-trained model, and injects small-scale trainable parameters for multiple downstream tasks (e.g., summarization, question answering and translation). To further enhance the efficiency of fine-tuning, we propose a framework that integrates LoRA and structured layer pruning. The integrated framework is validated on two created deidentified medical report summarization datasets based on MIMIC-IV-Note and two public medical dialogue datasets. By tuning 0.6% parameters of the original model and pruning over 30% Transformer-layers, our framework can reduce 50% of GPU memory usage and speed up 100% of the training phase, while preserving over 92% generation qualities on free-text sequence-to-sequence tasks.

Ding Li, Yongqiang Tang, Zhizhong Zhang et al.

Self-supervised skeleton-based action recognition enjoys a rapid growth along with the development of contrastive learning. The existing methods rely on imposing invariance to augmentations of 3D skeleton within a single data stream, which merely leverages the easy positive pairs and limits the ability to explore the complicated movement patterns. In this paper, we advocate that the defect of single-stream contrast and the lack of necessary feature transformation are responsible for easy positives, and therefore propose a Cross-Stream Contrastive Learning framework for skeleton-based action Representation learning (CSCLR). Specifically, the proposed CSCLR not only utilizes intra-stream contrast pairs, but introduces inter-stream contrast pairs as hard samples to formulate a better representation learning. Besides, to further exploit the potential of positive pairs and increase the robustness of self-supervised representation learning, we propose a Positive Feature Transformation (PFT) strategy which adopts feature-level manipulation to increase the variance of positive pairs. To validate the effectiveness of our method, we conduct extensive experiments on three benchmark datasets NTU-RGB+D 60, NTU-RGB+D 120 and PKU-MMD. Experimental results show that our proposed CSCLR exceeds the state-of-the-art methods on a diverse range of evaluation protocols.

Yajing Liu, Zhengya Sun, Wensheng Zhang

Graph neural networks (GNN) have emerged as a powerful tool for fraud detection tasks, where fraudulent nodes are identified by aggregating neighbor information via different relations. To get around such detection, crafty fraudsters resort to camouflage via connecting to legitimate users (i.e., relation camouflage) or providing seemingly legitimate feedbacks (i.e., feature camouflage). A wide-spread solution reinforces the GNN aggregation process with neighbor selectors according to original node features. This method may carry limitations when identifying fraudsters not only with the relation camouflage, but with the feature camouflage making them hard to distinguish from their legitimate neighbors. In this paper, we propose a Hierarchical Attention-based Graph Neural Network (HA-GNN) for fraud detection, which incorporates weighted adjacency matrices across different relations against camouflage. This is motivated in the Relational Density Theory and is exploited for forming a hierarchical attention-based graph neural network. Specifically, we design a relation attention module to reflect the tie strength between two nodes, while a neighborhood attention module to capture the long-range structural affinity associated with the graph. We generate node embeddings by aggregating information from local/long-range structures and original node features. Experiments on three real-world datasets demonstrate the effectiveness of our model over the state-of-the-arts.

Yunqi Zhu, Xuebing Yang, Yuanyuan Wu et al.

ROUGE is a standard automatic evaluation metric based on n-grams for sequence-to-sequence tasks, while cross-entropy loss is an essential objective of neural network language model that optimizes at a unigram level. We present differentiable n-gram objectives, attempting to alleviate the discrepancy between training criterion and evaluating criterion. The objective maximizes the probabilistic weight of matched sub-sequences, and the novelty of our work is the objective weights the matched sub-sequences equally and does not ceil the number of matched sub-sequences by the ground truth count of n-grams in reference sequence. We jointly optimize cross-entropy loss and the proposed objective, providing decent ROUGE score enhancement over abstractive summarization dataset CNN/DM and XSum, outperforming alternative n-gram objectives.

Zelin Ren, Xuebing Yang, Yuchen Jiang et al.

Kernel principal component analysis (KPCA) is a well-recognized nonlinear dimensionality reduction method that has been widely used in nonlinear fault detection tasks. As a kernel trick-based method, KPCA inherits two major problems. First, the form and the parameters of the kernel function are usually selected blindly, depending seriously on trial-and-error. As a result, there may be serious performance degradation in case of inappropriate selections. Second, at the online monitoring stage, KPCA has much computational burden and poor real-time performance, because the kernel method requires to leverage all the offline training data. In this work, to deal with the two drawbacks, a learnable faster realization of the conventional KPCA is proposed. The core idea is to parameterize all feasible kernel functions using the novel nonlinear DAE-FE (deep autoencoder based feature extraction) framework and propose DAE-PCA (deep autoencoder based principal component analysis) approach in detail. The proposed DAE-PCA method is proved to be equivalent to KPCA but has more advantage in terms of automatic searching of the most suitable nonlinear high-dimensional space according to the inputs. Furthermore, the online computational efficiency improves by approximately 100 times compared with the conventional KPCA. With the Tennessee Eastman (TE) process benchmark, the effectiveness and superiority of the proposed method is illustrated.

Ding Li, Rui Wu, Yongqiang Tang et al.

Video moment retrieval aims to search the moment most relevant to a given language query. However, most existing methods in this community often require temporal boundary annotations which are expensive and time-consuming to label. Hence weakly supervised methods have been put forward recently by only using coarse video-level label. Despite effectiveness, these methods usually process moment candidates independently, while ignoring a critical issue that the natural temporal dependencies between candidates in different temporal scales. To cope with this issue, we propose a Multi-scale 2D Representation Learning method for weakly supervised video moment retrieval. Specifically, we first construct a two-dimensional map for each temporal scale to capture the temporal dependencies between candidates. Two dimensions in this map indicate the start and end time points of these candidates. Then, we select top-K candidates from each scale-varied map with a learnable convolutional neural network. With a newly designed Moments Evaluation Module, we obtain the alignment scores of the selected candidates. At last, the similarity between captions and language query is served as supervision for further training the candidates' selector. Experiments on two benchmark datasets Charades-STA and ActivityNet Captions demonstrate that our approach achieves superior performance to state-of-the-art results.

Wensheng Zhang, Trent Muhr

This paper considers a federated learning system composed of a central coordinating server and multiple distributed local workers, all having access to trusted execution environments (TEEs). In order to ensure that the untrusted workers correctly perform local learning, we propose a new TEE-based approach that also combines techniques from applied cryptography, smart contract and game theory. Theoretical analysis and implementation-based evaluations show that, the proposed approach is secure, efficient and practical.

Wensheng Zhang, Yan Zheng, Taiga Miyazono et al.

Book covers are intentionally designed and provide an introduction to a book. However, they typically require professional skills to design and produce the cover images. Thus, we propose a generative neural network that can produce book covers based on an easy-to-use layout graph. The layout graph contains objects such as text, natural scene objects, and solid color spaces. This layout graph is embedded using a graph convolutional neural network and then used with a mask proposal generator and a bounding-box generator and filled using an object proposal generator. Next, the objects are compiled into a single image and the entire network is trained using a combination of adversarial training, perceptual training, and reconstruction. Finally, a Style Retention Network (SRNet) is used to transfer the learned font style onto the desired text. Using the proposed method allows for easily controlled and unique book covers.

Lei Yang, Zitong Zhang, Xiaotian Jia et al.

The Travelling Thief Problem (TTP) is a challenging combinatorial optimization problem that attracts many scholars. The TTP interconnects two well-known NP-hard problems: the Travelling Salesman Problem (TSP) and the 0-1 Knapsack Problem (KP). Increasingly algorithms have been proposed for solving this novel problem that combines two interdependent sub-problems. In this paper, TTP is investigated theoretically and empirically. An algorithm based on the score value calculated by our proposed formulation in picking items and sorting items in the reverse order in the light of the scoring value is proposed to solve the problem. Different approaches for solving the TTP are compared and analyzed; the experimental investigations suggest that our proposed approach is very efficient in meeting or beating current state-of-the-art heuristic solutions on a comprehensive set of benchmark TTP instances.

Yehui Yang, Fangxin Shang, Binghong Wu et al.

Diabetic retinopathy (DR) grading from fundus images has attracted increasing interest in both academic and industrial communities. Most convolutional neural network (CNN) based algorithms treat DR grading as a classification task via image-level annotations. However, these algorithms have not fully explored the valuable information in the DR-related lesions. In this paper, we present a robust framework, which collaboratively utilizes patch-level and image-level annotations, for DR severity grading. By an end-to-end optimization, this framework can bi-directionally exchange the fine-grained lesion and image-level grade information. As a result, it exploits more discriminative features for DR grading. The proposed framework shows better performance than the recent state-of-the-art algorithms and three clinical ophthalmologists with over nine years of experience. By testing on datasets of different distributions (such as label and camera), we prove that our algorithm is robust when facing image quality and distribution variations that commonly exist in real-world practice. We inspect the proposed framework through extensive ablation studies to indicate the effectiveness and necessity of each motivation. The code and some valuable annotations are now publicly available.

Liangchen Hu, Wensheng Zhang

With the dramatic increase of dimensions in the data representation, extracting latent low-dimensional features becomes of the utmost importance for efficient classification. Aiming at the problems of unclear margin representation and difficulty in revealing the data manifold structure in most of the existing linear discriminant methods, we propose a new discriminant feature extraction framework, namely Robust Locality-Aware Regression (RLAR). In our model, we introduce a retargeted regression to perform the marginal representation learning adaptively instead of using the general average inter-class margin. Besides, we formulate a new strategy for enhancing the local intra-class compactness of the data manifold, which can achieve the joint learning of locality-aware graph structure and desirable projection matrix. To alleviate the disturbance of outliers and prevent overfitting, we measure the regression term and locality-aware term together with the regularization term by the L2,1 norm. Further, forcing the row sparsity on the projection matrix through the L2,1 norm achieves the cooperation of feature selection and feature extraction. Then, we derive an effective iterative algorithm for solving the proposed model. The experimental results over a range of UCI data sets and other benchmark databases demonstrate that the proposed RLAR outperforms some state-of-the-art approaches.

Lei Tian, Yongqiang Tang, Liangchen Hu et al.

Domain adaptation has been a fundamental technology for transferring knowledge from a source domain to a target domain. The key issue of domain adaptation is how to reduce the distribution discrepancy between two domains in a proper way such that they can be treated indifferently for learning. In this paper, we propose a novel domain adaptation approach, which can thoroughly explore the data distribution structure of target domain.Specifically, we regard the samples within the same cluster in target domain as a whole rather than individuals and assigns pseudo-labels to the target cluster by class centroid matching. Besides, to exploit the manifold structure information of target data more thoroughly, we further introduce a local manifold self-learning strategy into our proposal to adaptively capture the inherent local connectivity of target samples. An efficient iterative optimization algorithm is designed to solve the objective function of our proposal with theoretical convergence guarantee. In addition to unsupervised domain adaptation, we further extend our method to the semi-supervised scenario including both homogeneous and heterogeneous settings in a direct but elegant way. Extensive experiments on seven benchmark datasets validate the significant superiority of our proposal in both unsupervised and semi-supervised manners.

Zewen He, He Huang, Yudong Wu et al.

Scale variation remains a challenging problem for object detection. Common paradigms usually adopt multiscale training & testing (image pyramid) or FPN (feature pyramid network) to process objects in a wide scale range. However, multi-scale methods aggravate more variations of scale that even deep convolution neural networks with FPN cannot handle well. In this work, we propose an innovative paradigm called Instance Scale Normalization (ISN) to resolve the above problem. ISN compresses the scale space of objects into a consistent range (ISN range), in both training and testing phases. This reassures the problem of scale variation fundamentally and reduces the difficulty of network optimization. Experiments show that ISN surpasses multi-scale counterpart significantly for object detection, instance segmentation, and multi-task human pose estimation, on several architectures. On COCO test-dev, our single model based on ISN achieves 46.5 mAP with a ResNet-101 backbone, which is among the state-of-the-art (SOTA) candidates for object detection.

JunPing Wang, WenSheng Zhang, Ian Thomas et al.

Generating sequential decision process from huge amounts of measured process data is a future research direction for collaborative factory automation, making full use of those online or offline process data to directly design flexible make decisions policy, and evaluate performance. The key challenges for the sequential decision process is to online generate sequential decision-making policy directly, and transferring knowledge across tasks domain. Most multi-task policy generating algorithms often suffer from insufficient generating cross-task sharing structure at discrete-time nonlinear systems with applications. This paper proposes the multi-task generative adversarial nets with shared memory for cross-domain coordination control, which can generate sequential decision policy directly from raw sensory input of all of tasks, and online evaluate performance of system actions in discrete-time nonlinear systems. Experiments have been undertaken using a professional flexible manufacturing testbed deployed within a smart factory of Weichai Power in China. Results on three groups of discrete-time nonlinear control tasks show that our proposed model can availably improve the performance of task with the help of other related tasks.

Haifang Li, Yingce Xia, Wensheng Zhang

Policy evaluation with linear function approximation is an important problem in reinforcement learning. When facing high-dimensional feature spaces, such a problem becomes extremely hard considering the computation efficiency and quality of approximations. We propose a new algorithm, LSTD($λ$)-RP, which leverages random projection techniques and takes eligibility traces into consideration to tackle the above two challenges. We carry out theoretical analysis of LSTD($λ$)-RP, and provide meaningful upper bounds of the estimation error, approximation error and total generalization error. These results demonstrate that LSTD($λ$)-RP can benefit from random projection and eligibility traces strategies, and LSTD($λ$)-RP can achieve better performances than prior LSTD-RP and LSTD($λ$) algorithms.

Zhizhong Zhang, Yuan Xie, Wensheng Zhang et al.

Multi-index fusion has demonstrated impressive performances in retrieval task by integrating different visual representations in a unified framework. However, previous works mainly consider propagating similarities via neighbor structure, ignoring the high order information among different visual representations. In this paper, we propose a new multi-index fusion scheme for image retrieval. By formulating this procedure as a multilinear based optimization problem, the complementary information hidden in different indexes can be explored more thoroughly. Specially, we first build our multiple indexes from various visual representations. Then a so-called index-specific functional matrix, which aims to propagate similarities, is introduced for updating the original index. The functional matrices are then optimized in a unified tensor space to achieve a refinement, such that the relevant images can be pushed more closer. The optimization problem can be efficiently solved by the augmented Lagrangian method with theoretical convergence guarantee. Unlike the traditional multi-index fusion scheme, our approach embeds the multi-index subspace structure into the new indexes with sparse constraint, thus it has little additional memory consumption in online query stage. Experimental evaluation on three benchmark datasets reveals that the proposed approach achieves the state-of-the-art performance, i.e., N-score 3.94 on UKBench, mAP 94.1\% on Holiday and 62.39\% on Market-1501.

Yanyun Qu, Jinyan Liu, Yuan Xie et al.

Most recently, tensor-SVD is implemented on multi-view self-representation clustering and has achieved the promising results in many real-world applications such as face clustering, scene clustering and generic object clustering. However, tensor-SVD based multi-view self-representation clustering is proposed originally to solve the clustering problem in the multiple linear subspaces, leading to unsatisfactory results when dealing with the case of non-linear subspaces. To handle data clustering from the non-linear subspaces, a kernelization method is designed by mapping the data from the original input space to a new feature space in which the transformed data can be clustered by a multiple linear clustering method. In this paper, we make an optimization model for the kernelized multi-view self-representation clustering problem. We also develop a new efficient algorithm based on the alternation direction method and infer a closed-form solution. Since all the subproblems can be solved exactly, the proposed optimization algorithm is guaranteed to obtain the optimal solution. In particular, the original tensor-based multi-view self-representation clustering problem is a special case of our approach and can be solved by our algorithm. Experimental results on several popular real-world clustering datasets demonstrate that our approach achieves the state-of-the-art performance.

Yehui Yang, Tao Li, Wensi Li et al.

We propose an automatic diabetic retinopathy (DR) analysis algorithm based on two-stages deep convolutional neural networks (DCNN). Compared to existing DCNN-based DR detection methods, the proposed algorithm have the following advantages: (1) Our method can point out the location and type of lesions in the fundus images, as well as giving the severity grades of DR. Moreover, since retina lesions and DR severity appear with different scales in fundus images, the integration of both local and global networks learn more complete and specific features for DR analysis. (2) By introducing imbalanced weighting map, more attentions will be given to lesion patches for DR grading, which significantly improve the performance of the proposed algorithm. In this study, we label 12,206 lesion patches and re-annotate the DR grades of 23,595 fundus images from Kaggle competition dataset. Under the guidance of clinical ophthalmologists, the experimental results show that our local lesion detection net achieve comparable performance with trained human observers, and the proposed imbalanced weighted scheme also be proved to significantly improve the capability of our DCNN-based DR grading algorithm.

Yuan Xie, Dacheng Tao, Wensheng Zhang et al.

In this paper, we address the multi-view subspace clustering problem. Our method utilizes the circulant algebra for tensor, which is constructed by stacking the subspace representation matrices of different views and then rotating, to capture the low rank tensor subspace so that the refinement of the view-specific subspaces can be achieved, as well as the high order correlations underlying multi-view data can be explored.} By introducing a recently proposed tensor factorization, namely tensor-Singular Value Decomposition (t-SVD) \cite{kilmer13}, we can impose a new type of low-rank tensor constraint on the rotated tensor to capture the complementary information from multiple views. Different from traditional unfolding based tensor norm, this low-rank tensor constraint has optimality properties similar to that of matrix rank derived from SVD, so the complementary information among views can be explored more efficiently and thoroughly. The established model, called t-SVD based Multi-view Subspace Clustering (t-SVD-MSC), falls into the applicable scope of augmented Lagrangian method, and its minimization problem can be efficiently solved with theoretical convergence guarantee and relatively low computational complexity. Extensive experimental testing on eight challenging image dataset shows that the proposed method has achieved highly competent objective performance compared to several state-of-the-art multi-view clustering methods.

Yuan Xie, Shuhang Gu, Yan Liu et al.

Low rank matrix approximation (LRMA), which aims to recover the underlying low rank matrix from its degraded observation, has a wide range of applications in computer vision. The latest LRMA methods resort to using the nuclear norm minimization (NNM) as a convex relaxation of the nonconvex rank minimization. However, NNM tends to over-shrink the rank components and treats the different rank components equally, limiting its flexibility in practical applications. We propose a more flexible model, namely the Weighted Schatten $p$-Norm Minimization (WSNM), to generalize the NNM to the Schatten $p$-norm minimization with weights assigned to different singular values. The proposed WSNM not only gives better approximation to the original low-rank assumption, but also considers the importance of different rank components. We analyze the solution of WSNM and prove that, under certain weights permutation, WSNM can be equivalently transformed into independent non-convex $l_p$-norm subproblems, whose global optimum can be efficiently solved by generalized iterated shrinkage algorithm. We apply WSNM to typical low-level vision problems, e.g., image denoising and background subtraction. Extensive experimental results show, both qualitatively and quantitatively, that the proposed WSNM can more effectively remove noise, and model complex and dynamic scenes compared with state-of-the-art methods.

Wenrui Hu, Dacheng Tao, Wensheng Zhang et al.

In this paper, we propose a new low-rank tensor model based on the circulant algebra, namely, twist tensor nuclear norm or t-TNN for short. The twist tensor denotes a 3-way tensor representation to laterally store 2D data slices in order. On one hand, t-TNN convexly relaxes the tensor multi-rank of the twist tensor in the Fourier domain, which allows an efficient computation using FFT. On the other, t-TNN is equal to the nuclear norm of block circulant matricization of the twist tensor in the original domain, which extends the traditional matrix nuclear norm in a block circulant way. We test the t-TNN model on a video completion application that aims to fill missing values and the experiment results validate its effectiveness, especially when dealing with video recorded by a non-stationary panning camera. The block circulant matricization of the twist tensor can be transformed into a circulant block representation with nuclear norm invariance. This representation, after transformation, exploits the horizontal translation relationship between the frames in a video, and endows the t-TNN model with a more powerful ability to reconstruct panning videos than the existing state-of-the-art low-rank models.

Yuan Xie, Wensheng Zhang, Dacheng Tao et al.

It remains a challenge to simultaneously remove geometric distortion and space-time-varying blur in frames captured through a turbulent atmospheric medium. To solve, or at least reduce these effects, we propose a new scheme to recover a latent image from observed frames by integrating a new variational model and distortion-driven spatial-temporal kernel regression. The proposed scheme first constructs a high-quality reference image from the observed frames using low-rank decomposition. Then, to generate an improved registered sequence, the reference image is iteratively optimized using a variational model containing a new spatial-temporal regularization. The proposed fast algorithm efficiently solves this model without the use of partial differential equations (PDEs). Next, to reduce blur variation, distortion-driven spatial-temporal kernel regression is carried out to fuse the registered sequence into one image by introducing the concept of the near-stationary patch. Applying a blind deconvolution algorithm to the fused image produces the final output. Extensive experimental testing shows, both qualitatively and quantitatively, that the proposed method can effectively alleviate distortion and blur and recover details of the original scene compared to state-of-the-art methods.