arXiv subjects

Wanli Ma

Publications and source records attributed to Wanli Ma.

30 records · Page 2Linked to original sources

arXiv2025

Relation Also Knows: Rethinking the Recall and Editing of Factual Associations in Auto-Regressive Transformer Language Models

The storage and recall of factual associations in auto-regressive transformer language models (LMs) have drawn a great deal of attention, inspiring knowledge editing by directly modifying the located model weights. Most editing works achieve knowledge editing under the guidance of existing interpretations of knowledge recall that mainly focus on subject knowledge. However, these interpretations are seriously flawed, neglecting relation information and leading to the over-generalizing problem for editing. In this work, we discover a novel relation-focused perspective to interpret the knowledge recall of transformer LMs during inference and apply it on single knowledge editing to avoid over-generalizing. Experimental results on the dataset supplemented with a new R-Specificity criterion demonstrate that our editing approach significantly alleviates over-generalizing while remaining competitive on other criteria, breaking the domination of subject-focused editing for future research.

cs.CL↗

arXiv2025

Patch-GAN Transfer Learning with Reconstructive Models for Cloud Removal

Cloud removal plays a crucial role in enhancing remote sensing image analysis, yet accurately reconstructing cloud-obscured regions remains a significant challenge. Recent advancements in generative models have made the generation of realistic images increasingly accessible, offering new opportunities for this task. Given the conceptual alignment between image generation and cloud removal tasks, generative models present a promising approach for addressing cloud removal in remote sensing. In this work, we propose a deep transfer learning approach built on a generative adversarial network (GAN) framework to explore the potential of the novel masked autoencoder (MAE) image reconstruction model in cloud removal. Due to the complexity of remote sensing imagery, we further propose using a patch-wise discriminator to determine whether each patch of the image is real or not. The proposed reconstructive transfer learning approach demonstrates significant improvements in cloud removal performance compared to other GAN-based methods. Additionally, whilst direct comparisons with some of the state-of-the-art cloud removal techniques are limited due to unclear details regarding their train/test data splits, the proposed model achieves competitive results based on available benchmarks.

cs.CV↗

arXiv2024

A Survey on the Principles of Persuasion as a Social Engineering Strategy in Phishing

Research shows that phishing emails often utilize persuasion techniques, such as social proof, liking, consistency, authority, scarcity, and reciprocity to gain trust to obtain sensitive information or maliciously infect devices. The link between principles of persuasion and social engineering attacks, particularly in phishing email attacks, is an important topic in cyber security as they are the common and effective method used by cybercriminals to obtain sensitive information or access computer systems. This survey paper concluded that spear phishing, a targeted form of phishing, has been found to be specifically effective as attackers can tailor their messages to the specific characteristics, interests, and vulnerabilities of their targets. Understanding the uses of the principles of persuasion in spear phishing is key to the effective defence against it and eventually its elimination. This survey paper systematically summarizes and presents the current state of the art in understanding the use of principles of persuasion in phishing. Through a systematic review of the existing literature, this survey paper identifies a significant gap in the understanding of the impact of principles of persuasion as a social engineering strategy in phishing attacks and highlights the need for further research in this area.

cs.CR↗

arXiv2024

Understanding Auditory Evoked Brain Signal via Physics-informed Embedding Network with Multi-Task Transformer

In the fields of brain-computer interaction and cognitive neuroscience, effective decoding of auditory signals from task-based functional magnetic resonance imaging (fMRI) is key to understanding how the brain processes complex auditory information. Although existing methods have enhanced decoding capabilities, limitations remain in information utilization and model representation. To overcome these challenges, we propose an innovative multi-task learning model, Physics-informed Embedding Network with Multi-Task Transformer (PEMT-Net), which enhances decoding performance through physics-informed embedding and deep learning techniques. PEMT-Net consists of two principal components: feature augmentation and classification. For feature augmentation, we propose a novel approach by creating neural embedding graphs via node embedding, utilizing random walks to simulate the physical diffusion of neural information. This method captures both local and non-local information overflow and proposes a position encoding based on relative physical coordinates. In the classification segment, we propose adaptive embedding fusion to maximally capture linear and non-linear characteristics. Furthermore, we propose an innovative parameter-sharing mechanism to optimize the retention and learning of extracted features. Experiments on a specific dataset demonstrate PEMT-Net's significant performance in multi-task auditory signal decoding, surpassing existing methods and offering new insights into the brain's mechanisms for processing complex auditory information.

q-bio.NC↗

arXiv2024

Knowledge Distillation for Road Detection based on cross-model Semi-Supervised Learning

The advancement of knowledge distillation has played a crucial role in enabling the transfer of knowledge from larger teacher models to smaller and more efficient student models, and is particularly beneficial for online and resource-constrained applications. The effectiveness of the student model heavily relies on the quality of the distilled knowledge received from the teacher. Given the accessibility of unlabelled remote sensing data, semi-supervised learning has become a prevalent strategy for enhancing model performance. However, relying solely on semi-supervised learning with smaller models may be insufficient due to their limited capacity for feature extraction. This limitation restricts their ability to exploit training data. To address this issue, we propose an integrated approach that combines knowledge distillation and semi-supervised learning methods. This hybrid approach leverages the robust capabilities of large models to effectively utilise large unlabelled data whilst subsequently providing the small student model with rich and informative features for enhancement. The proposed semi-supervised learning-based knowledge distillation (SSLKD) approach demonstrates a notable improvement in the performance of the student model, in the application of road segmentation, surpassing the effectiveness of traditional semi-supervised learning methods.

cs.CV↗

arXiv2023

Confidence-Guided Semi-supervised Learning in Land Cover Classification

Semi-supervised learning has been well developed to help reduce the cost of manual labelling by exploiting a large quantity of unlabelled data. Especially in the application of land cover classification, pixel-level manual labelling in large-scale imagery is labour-intensive, time-consuming and expensive. However, existing semi-supervised learning methods pay limited attention to the quality of pseudo-labels during training even though the quality of training data is one of the critical factors determining network performance. In order to fill this gap, we develop a confidence-guided semi-supervised learning (CGSSL) approach to make use of high-confidence pseudo labels and reduce the negative effect of low-confidence ones for land cover classification. Meanwhile, the proposed semi-supervised learning approach uses multiple network architectures to increase the diversity of pseudo labels. The proposed semi-supervised learning approach significantly improves the performance of land cover classification compared to the classic semi-supervised learning methods and even outperforms fully supervised learning with a complete set of labelled imagery of the benchmark Potsdam land cover dataset.

cs.CV↗

arXiv2023

Noda Iteration for Computing Generalized Tensor Eigenpairs

In this paper, we propose the tensor Noda iteration (NI) and its inexact version for solving the eigenvalue problem of a particular class of tensor pairs called generalized $\mathcal{M}$-tensor pairs. A generalized $\mathcal{M}$-tensor pair consists of a weakly irreducible nonnegative tensor and a nonsingular $\mathcal{M}$-tensor within a linear combination. It is shown that any generalized $\mathcal{M}$-tensor pair admits a unique positive generalized eigenvalue with a positive eigenvector. A modified tensor Noda iteration(MTNI) is developed for extending the Noda iteration for nonnegative matrix eigenproblems. In addition, the inexact generalized tensor Noda iteration method (IGTNI) and the generalized Newton-Noda iteration method (GNNI) are also introduced for more efficient implementations and faster convergence. Under a mild assumption on the initial values, the convergence of these algorithms is guaranteed. The efficiency of these algorithms is illustrated by numerical experiments.

math.NA↗

arXiv2022

Reinforcement Learning based Voice Interaction to Clear Path for Robots in Elevator Environment

Efficient use of the space in an elevator is very necessary for a service robot, due to the need for reducing the amount of time caused by waiting for the next elevator. To provide a solution for this, we propose a hybrid approach that combines reinforcement learning (RL) with voice interaction for robot navigation in the scene of entering the elevator. RL provides robots with a high exploration ability to find a new clear path to enter the elevator compared to traditional navigation methods such as Optimal Reciprocal Collision Avoidance (ORCA). The proposed method allows the robot to take an active clear path action towards the elevator whilst a crowd of people stands at the entrance of the elevator wherein there are still lots of space. This is done by embedding a clear path action (voice prompt) into the RL framework, and the proposed navigation policy helps the robot to finish tasks efficiently and safely. Our model approach provides a great improvement in the success rate and reward of entering the elevator compared to state-of-the-art navigation policies without active clear path operation.

cs.RO↗

arXiv2022

High-precision Density Mapping of Marine Debris and Floating Plastics via Satellite Imagery

Combining multi-spectral satellite data and machine learning has been suggested as a method for monitoring plastic pollutants in the ocean environment. Recent studies have made theoretical progress regarding the identification of marine plastic via machine learning. However, no study has assessed the application of these methods for mapping and monitoring marine-plastic density. As such, this paper comprised of three main components: (1) the development of a machine learning model, (2) the construction of the MAP-Mapper, an automated tool for mapping marine-plastic density, and finally (3) an evaluation of the whole system for out-of-distribution test locations. The findings from this paper leverage the fact that machine learning models need to be high-precision to reduce the impact of false positives on results. The developed MAP-Mapper architectures provide users choices to reach high-precision ($\textit{abbv.}$ -HP) or optimum precision-recall ($\textit{abbv.}$ -Opt) values in terms of the training/test data set. Our MAP-Mapper-HP model greatly increased the precision of plastic detection to 95\%, whilst MAP-Mapper-Opt reaches precision-recall pair of 87\%-88\%. The MAP-Mapper contributes to the literature with the first tool to exploit advanced deep/machine learning and multi-spectral imagery to map marine-plastic density in automated software. The proposed data pipeline has taken a novel approach to map plastic density in ocean regions. As such, this enables an initial assessment of the challenges and opportunities of this method to help guide future work and scientific study.

eess.IV↗

arXiv2020

Exploiting the Dual-Tree Complex Wavelet Transform for Ship Wake Detection in SAR Imagery

In this paper, we analyse synthetic aperture radar (SAR) images of the sea surface using an inverse problem formulation whereby Radon domain information is enhanced in order to accurately detect ship wakes. This is achieved by promoting linear features in the images. For the inverse problem-solving stage, we propose a penalty function, which combines the dual-tree complex wavelet transform (DT-CWT) with the non-convex Cauchy penalty function. The solution to this inverse problem is based on the forward-backward (FB) splitting algorithm to obtain enhanced images in the Radon domain. The proposed method achieves the best results and leads to significant improvement in terms of various performance metrics, compared to state-of-the-art ship wake detection methods. The accuracy of detecting ship wakes in SAR images with different frequency bands and spatial resolution reaches more than 90%, which clearly demonstrates an accuracy gain of 7% compared to the second-best approach.

eess.IV↗

arXiv2019

Evanescent Field Functional Cu3-xP Nanoparticles as Effective Saturable Absorbers with high Repeatability for Femtosecond Soliton Pulse generation

Recently, a new emerging field about heavily-doped colloidal plasmonic nanocrystals (NCs) has attracted great attention due to their lower and expediently adjustable free carrier densities, lower and tunable LSPR band in the spectral range from NIR to MIR and higher optical nonlinearity. These new kinds of plasmonic materials will show huge potential and opportunities for nonlinear optical applications, such as ultrafast switching, nonlinear sensing and pulse laser generation. In this work, we demonstrate that high-quality mode-locking and Q-switching pulses at 1560 nm can both be generated by using controllable concentration of Cu3-xP NCs solution and fabricating evanescently interacted saturable absorbers. Furthermore, the plasmonic NCs material has good reproduction for fabricating SA devices and promising potential for large-scale industrial production. Our results may attract great attention for further investigations of heavily-doped plasmonic NCs as next generation, cheap and solution-processed element for fascinating applications in optoelectronic devices.

physics.app-ph↗

arXiv2015

Efficient Structured Matrix Rank Minimization

We study the problem of finding structured low-rank matrices using nuclear norm regularization where the structure is encoded by a linear map. In contrast to most known approaches for linearly structured rank minimization, we do not (a) use the full SVD, nor (b) resort to augmented Lagrangian techniques, nor (c) solve linear systems per iteration. Instead, we formulate the problem differently so that it is amenable to a generalized conditional gradient method, which results in a practical improvement with low per iteration computational cost. Numerical results show that our approach significantly outperforms state-of-the-art competitors in terms of running time, while effectively recovering low rank solutions in stochastic system realization and spectral compressed sensing problems.

eess.SY↗