Abstract
Multivariate Time Series (MTS) data capture temporal behaviors to provide invaluable insights into various physical dynamic phenomena. In smart mobility, MTS plays a crucial role in providing temporal dynamics of behaviors such as maneuver patterns, enabling early detection of anomalous behaviors while facilitating pro-activity in Prognostics and Health Management (PHM). In this work, we aim to address challenges associated with modeling MTS data collected from a vehicle using sensors. Our goal is to investigate the effectiveness of two distinct unsupervised representation learning approaches in identifying maneuvering states in smart mobility. Specifically, we focus on some bivariate accelerations extracted from 2.5 years of driving, where the dataset is non-stationary, long, noisy, and completely unlabeled, making manual labeling impractical. The approaches of interest are Temporal Neighborhood Coding for Maneuvering (TNC4Maneuvering) and Decoupled Local and Global Representation learner for Maneuvering (DLG4Maneuvering). The main advantage of these frameworks is that they capture transferable insights in a form of representations from the data that can be effectively applied in multiple subsequent tasks, such as time-series classification, clustering, and multi-linear regression, which are the quantitative measures and qualitative measures, including visualization of representations themselves and resulting reconstructed MTS, respectively. We compare their effectiveness, where possible, in order to gain insights into which approach is more effective in identifying maneuvering states in smart mobility.
Title:
Cross-Modal Self-Supervised Learning with Effective Contrastive Units for LiDAR Point Clouds
Authors: Mu Cai, Chenxu Luo, Yong Jae Lee, Xiaodong Yang
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
3D perception in LiDAR point clouds is crucial for a self-driving vehicle to properly act in 3D environment. However, manually labeling point clouds is hard and costly. There has been a growing interest in self-supervised pre-training of 3D perception models. Following the success of contrastive learning in images, current methods mostly conduct contrastive pre-training on point clouds only. Yet an autonomous driving vehicle is typically supplied with multiple sensors including cameras and LiDAR. In this context, we systematically study single modality, cross-modality, and multi-modality for contrastive learning of point clouds, and show that cross-modality wins over other alternatives. In addition, considering the huge difference between the training sources in 2D images and 3D point clouds, it remains unclear how to design more effective contrastive units for LiDAR. We therefore propose the instance-aware and similarity-balanced contrastive units that are tailored for self-driving point clouds. Extensive experiments reveal that our approach achieves remarkable performance gains over various point cloud models across the downstream perception tasks of LiDAR based 3D object detection and 3D semantic segmentation on the four popular benchmarks including Waymo Open Dataset, nuScenes, SemanticKITTI and ONCE.
Title:
Atom dimension adaptation for infinite set dictionary learning
Authors: Andra Băltoiu, Denis C. Ilie-Ablachim, Bogdan Dumitrescu
Abstract
Recent work on dictionary learning with set-atoms has shown benefits in anomaly detection. Instead of viewing an atom as a single vector, these methods allow building sparse representations with atoms taken from a set around a central vector; the set can be a cone or may have a probability distribution associated to it. We propose a method for adaptively adjusting the size of set-atoms in Gaussian and cone dictionary learning. The purpose of the algorithm is to match the atom sizes with their contribution in representing the signals. The proposed algorithm not only decreases the representation error, but also improves anomaly detection, for a class of anomalies called `dependency'. We obtain better detection performance than state-of-the-art methods.
Title:
Few-Shot Learning: Expanding ID Cards Presentation Attack Detection to Unknown ID Countries
Authors: Alvaro S. Rocamora, Juan M. Espin, Juan E. Tapia
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
This paper proposes a Few-shot Learning (FSL) approach for detecting Presentation Attacks on ID Cards deployed in a remote verification system and its extension to new countries. Our research analyses the performance of Prototypical Networks across documents from Spain and Chile as a baseline and measures the extension of generalisation capabilities of new ID Card countries such as Argentina and Costa Rica. Specifically targeting the challenge of screen display presentation attacks. By leveraging convolutional architectures and meta-learning principles embodied in Prototypical Networks, we have crafted a model that demonstrates high efficacy with Few-shot examples. This research reveals that competitive performance can be achieved with as Few-shots as five unique identities and with under 100 images per new country added. This opens a new insight for novel generalised Presentation Attack Detection on ID cards to unknown attacks.
Title:
AssistTaxi: A Comprehensive Dataset for Taxiway Analysis and Autonomous Operations
Authors: Parth Ganeriwala, Siddhartha Bhattacharyya, Sean Gunther, Brian Kish, Mohammed Abdul Hafeez Khan, Ankur Dhadoti, Natasha Neogi
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
The availability of high-quality datasets play a crucial role in advancing research and development especially, for safety critical and autonomous systems. In this paper, we present AssistTaxi, a comprehensive novel dataset which is a collection of images for runway and taxiway analysis. The dataset comprises of more than 300,000 frames of diverse and carefully collected data, gathered from Melbourne (MLB) and Grant-Valkaria (X59) general aviation airports. The importance of AssistTaxi lies in its potential to advance autonomous operations, enabling researchers and developers to train and evaluate algorithms for efficient and safe taxiing. Researchers can utilize AssistTaxi to benchmark their algorithms, assess performance, and explore novel approaches for runway and taxiway analysis. Addition-ally, the dataset serves as a valuable resource for validating and enhancing existing algorithms, facilitating innovation in autonomous operations for aviation. We also propose an initial approach to label the dataset using a contour based detection and line extraction technique.
Title:
Mazed and Confused: A Dataset of Cybersickness, Working Memory, Mental Load, Physical Load, and Attention During a Real Walking Task in VR
Authors: Jyotirmay Nag Setu, Joshua M Le, Ripan Kumar Kundu, Barry Giesbrecht, Tobias Höllerer, Khaza Anuarul Hoque, Kevin Desai, John Quarles
Abstract
Virtual Reality (VR) is quickly establishing itself in various industries, including training, education, medicine, and entertainment, in which users are frequently required to carry out multiple complex cognitive and physical activities. However, the relationship between cognitive activities, physical activities, and familiar feelings of cybersickness is not well understood and thus can be unpredictable for developers. Researchers have previously provided labeled datasets for predicting cybersickness while users are stationary, but there have been few labeled datasets on cybersickness while users are physically walking. Thus, from 39 participants, we collected head orientation, head position, eye tracking, images, physiological readings from external sensors, and the self-reported cybersickness severity, physical load, and mental load in VR. Throughout the data collection, participants navigated mazes via real walking and performed tasks challenging their attention and working memory. To demonstrate the dataset's utility, we conducted a case study of training classifiers in which we achieved 95% accuracy for cybersickness severity classification. The noteworthy performance of the straightforward classifiers makes this dataset ideal for future researchers to develop cybersickness detection and reduction models. To better understand the features that helped with classification, we performed SHAP(SHapley Additive exPlanations) analysis, highlighting the importance of eye tracking and physiological measures for cybersickness prediction while walking. This open dataset can allow future researchers to study the connection between cybersickness and cognitive loads and develop prediction models. This dataset will empower future VR developers to design efficient and effective Virtual Environments by improving cognitive load management and minimizing cybersickness.
Title:
FSMDet: Vision-guided feature diffusion for fully sparse 3D detector
Authors: Tianran Liu, Morteza Mousa Pasandi, Robert Laganiere
Abstract
Fully sparse 3D detection has attracted an increasing interest in the recent years. However, the sparsity of the features in these frameworks challenges the generation of proposals because of the limited diffusion process. In addition, the quest for efficiency has led to only few work on vision-assisted fully sparse models. In this paper, we propose FSMDet (Fully Sparse Multi-modal Detection), which use visual information to guide the LiDAR feature diffusion process while still maintaining the efficiency of the pipeline. Specifically, most of fully sparse works focus on complex customized center fusion diffusion/regression operators. However, we observed that if the adequate object completion is performed, even the simplest interpolation operator leads to satisfactory results. Inspired by this observation, we split the vision-guided diffusion process into two modules: a Shape Recover Layer (SRLayer) and a Self Diffusion Layer (SDLayer). The former uses RGB information to recover the shape of the visible part of an object, and the latter uses a visual prior to further spread the features to the center region. Experiments demonstrate that our approach successfully improves the performance of previous fully sparse models that use LiDAR only and reaches SOTA performance in multimodal models. At the same time, thanks to the sparse architecture, our method can be up to 5 times more efficient than previous SOTA methods in the inference process.
Title:
Equivariant Filter for Tightly Coupled LiDAR-Inertial Odometry
Authors: Anbo Tao, Yarong Luo, Chunxi Xia, Chi Guo, Xingxing Li
Subjects: Subjects:
Robotics (cs.RO); Systems and Control (eess.SY)
Abstract
Pose estimation is a crucial problem in simultaneous localization and mapping (SLAM). However, developing a robust and consistent state estimator remains a significant challenge, as the traditional extended Kalman filter (EKF) struggles to handle the model nonlinearity, especially for inertial measurement unit (IMU) and light detection and ranging (LiDAR). To provide a consistent and efficient solution of pose estimation, we propose Eq-LIO, a robust state estimator for tightly coupled LIO systems based on an equivariant filter (EqF). Compared with the invariant Kalman filter based on the $\SE_2(3)$ group structure, the EqF uses the symmetry of the semi-direct product group to couple the system state including IMU bias, navigation state and LiDAR extrinsic calibration state, thereby suppressing linearization error and improving the behavior of the estimator in the event of unexpected state changes. The proposed Eq-LIO owns natural consistency and higher robustness, which is theoretically proven with mathematical derivation and experimentally verified through a series of tests on both public and private datasets.
Title:
Brain-Inspired Stepwise Patch Merging for Vision Transformers
Abstract
The hierarchical architecture has become a mainstream design paradigm for Vision Transformers (ViTs), with Patch Merging serving as the pivotal component that transforms a columnar architecture into a hierarchical one. Drawing inspiration from the brain's ability to integrate global and local information for comprehensive visual understanding, we propose a novel technique called Stepwise Patch Merging (SPM), which enhances the subsequent attention mechanism's ability to 'see' better. SPM comprises two critical modules: Multi-Scale Aggregation (MSA) and Guided Local Enhancement (GLE). The MSA module integrates multi-scale features to enrich feature representation, while the GLE module focuses on refining local detail extraction, thus achieving an optimal balance between long-range dependency modeling and local feature enhancement. Extensive experiments conducted on benchmark datasets, including ImageNet-1K, COCO, and ADE20K, demonstrate that SPM significantly improves the performance of various models, particularly in dense prediction tasks such as object detection and semantic segmentation. These results underscore the efficacy of SPM in enhancing model accuracy and robustness across a wide range of computer vision tasks.
Abstract
The detection and localization of deepfake content, particularly when small fake segments are seamlessly mixed with real videos, remains a significant challenge in the field of digital media security. Based on the recently released AV-Deepfake1M dataset, which contains more than 1 million manipulated videos across more than 2,000 subjects, we introduce the 1M-Deepfakes Detection Challenge. This challenge is designed to engage the research community in developing advanced methods for detecting and localizing deepfake manipulations within the large-scale high-realistic audio-visual dataset. The participants can access the AV-Deepfake1M dataset and are required to submit their inference results for evaluation across the metrics for detection or localization tasks. The methodologies developed through the challenge will contribute to the development of next-generation deepfake detection and localization systems. Evaluation scripts, baseline models, and accompanying code will be available on this https URL.
Title:
Graph sub-sampling for divide-and-conquer algorithms in large networks
Authors: Eric Yanchenko
Subjects: Subjects:
Social and Information Networks (cs.SI); Computation (stat.CO)
Abstract
As networks continue to increase in size, current methods must be capable of handling large numbers of nodes and edges in order to be practically relevant. Instead of working directly with the entire (large) network, analyzing sub-networks has become a popular approach. Due to a network's inherent inter-connectedness, sub-sampling is not a trivial task. While this problem has gained attention in recent years, it has not received sufficient attention from the statistics community. In this work, we provide a thorough comparison of seven graph sub-sampling algorithms by applying them to divide-and-conquer algorithms for community structure and core-periphery (CP) structure. After discussing the various algorithms and sub-sampling routines, we derive theoretical results for the mis-classification rate of the divide-and-conquer algorithm for CP structure under various sub-sampling schemes. We then perform extensive experiments on both simulated and real-world data to compare the various methods. For the community detection task, we found that sampling nodes uniformly at random yields the best performance. For CP structure on the other hand, there was no single winner, but algorithms which sampled core nodes at a higher rate consistently outperformed other sampling routines, e.g., random edge sampling and random walk sampling. The varying performance of the sampling algorithms on different tasks demonstrates the importance of carefully selecting a sub-sampling routine for the specific application.
Title:
ODYSSEE: Oyster Detection Yielded by Sensor Systems on Edge Electronics
Authors: Xiaomin Lin, Vivek Mange, Arjun Suresh, Bernhard Neuberger, Aadi Palnitkar, Brendan Campbell, Alan Williams, Kleio Baxevani, Jeremy Mallette, Alhim Vera, Markus Vincze, Ioannis Rekleitis, Herbert G. Tanner, Yiannis Aloimonos
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
Abstract
Oysters are a keystone species in coastal ecosystems, offering significant economic, environmental, and cultural benefits. However, current monitoring systems are often destructive, typically involving dredging to physically collect and count oysters. A nondestructive alternative is manual identification from video footage collected by divers, which is time-consuming and labor-intensive with expert input. An alternative to human monitoring is the deployment of a system with trained object detection models that performs real-time, on edge oyster detection in the field. One such platform is the Aqua2 robot. Effective training of these models requires extensive high-quality data, which is difficult to obtain in marine settings. To address these complications, we introduce a novel method that leverages stable diffusion to generate high-quality synthetic data for the marine domain. We exploit diffusion models to create photorealistic marine imagery, using ControlNet inputs to ensure consistency with the segmentation ground-truth mask, the geometry of the scene, and the target domain of real underwater images for oysters. The resulting dataset is used to train a YOLOv10-based vision model, achieving a state-of-the-art 0.657 mAP@50 for oyster detection on the Aqua2 platform. The system we introduce not only improves oyster habitat monitoring, but also paves the way to autonomous surveillance for various tasks in marine contexts, improving aquaculture and conservation efforts.
Title:
Improving Anomalous Sound Detection via Low-Rank Adaptation Fine-Tuning of Pre-Trained Audio Models
Abstract
Anomalous Sound Detection (ASD) has gained significant interest through the application of various Artificial Intelligence (AI) technologies in industrial settings. Though possessing great potential, ASD systems can hardly be readily deployed in real production sites due to the generalization problem, which is primarily caused by the difficulty of data collection and the complexity of environmental factors. This paper introduces a robust ASD model that leverages audio pre-trained models. Specifically, we fine-tune these models using machine operation data, employing SpecAug as a data augmentation strategy. Additionally, we investigate the impact of utilizing Low-Rank Adaptation (LoRA) tuning instead of full fine-tuning to address the problem of limited data for fine-tuning. Our experiments on the DCASE2023 Task 2 dataset establish a new benchmark of 77.75% on the evaluation set, with a significant improvement of 6.48% compared with previous state-of-the-art (SOTA) models, including top-tier traditional convolutional networks and speech pre-trained models, which demonstrates the effectiveness of audio pre-trained models with LoRA tuning. Ablation studies are also conducted to showcase the efficacy of the proposed scheme.
Title:
SCLNet: A Scale-Robust Complementary Learning Network for Object Detection in UAV Images
Authors: Xuexue Li
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Most recent UAV (Unmanned Aerial Vehicle) detectors focus primarily on general challenge such as uneven distribution and occlusion. However, the neglect of scale challenges, which encompass scale variation and small objects, continues to hinder object detection in UAV images. Although existing works propose solutions, they are implicitly modeled and have redundant steps, so detection performance remains limited. And one specific work addressing the above scale challenges can help improve the performance of UAV image detectors. Compared to natural scenes, scale challenges in UAV images happen with problems of limited perception in comprehensive scales and poor robustness to small objects. We found that complementary learning is beneficial for the detection model to address the scale challenges. Therefore, the paper introduces it to form our scale-robust complementary learning network (SCLNet) in conjunction with the object detection model. The SCLNet consists of two implementations and a cooperation method. In detail, one implementation is based on our proposed scale-complementary decoder and scale-complementary loss function to explicitly extract complementary information as complement, named comprehensive-scale complementary learning (CSCL). Another implementation is based on our proposed contrastive complement network and contrastive complement loss function to explicitly guide the learning of small objects with the rich texture detail information of the large objects, named inter-scale contrastive complementary learning (ICCL). In addition, an end-to-end cooperation (ECoop) between two implementations and with the detection model is proposed to exploit each potential.
Title:
A Continual and Incremental Learning Approach for TinyML On-device Training Using Dataset Distillation and Model Size Adaption
Authors: Marcus Rüb, Philipp Tuchel, Axel Sikora, Daniel Mueller-Gritschneder
Abstract
A new algorithm for incremental learning in the context of Tiny Machine learning (TinyML) is presented, which is optimized for low-performance and energy efficient embedded devices. TinyML is an emerging field that deploys machine learning models on resource-constrained devices such as microcontrollers, enabling intelligent applications like voice recognition, anomaly detection, predictive maintenance, and sensor data processing in environments where traditional machine learning models are not feasible. The algorithm solve the challenge of catastrophic forgetting through the use of knowledge distillation to create a small, distilled dataset. The novelty of the method is that the size of the model can be adjusted dynamically, so that the complexity of the model can be adapted to the requirements of the task. This offers a solution for incremental learning in resource-constrained environments, where both model size and computational efficiency are critical factors. Results show that the proposed algorithm offers a promising approach for TinyML incremental learning on embedded devices. The algorithm was tested on five datasets including: CIFAR10, MNIST, CORE50, HAR, Speech Commands. The findings indicated that, despite using only 43% of Floating Point Operations (FLOPs) compared to a larger fixed model, the algorithm experienced a negligible accuracy loss of just 1%. In addition, the presented method is memory efficient. While state-of-the-art incremental learning is usually very memory intensive, the method requires only 1% of the original data set.
Title:
Unsupervised Novelty Detection Methods Benchmarking with Wavelet Decomposition
Abstract
Novelty detection is a critical task in various engineering fields. Numerous approaches to novelty detection rely on supervised or semi-supervised learning, which requires labelled datasets for training. However, acquiring labelled data, when feasible, can be expensive and time-consuming. For these reasons, unsupervised learning is a powerful alternative that allows performing novelty detection without needing labelled samples. In this study, numerous unsupervised machine learning algorithms for novelty detection are compared, highlighting their strengths and weaknesses in the context of vibration sensing. The proposed framework uses a continuous metric, unlike most traditional methods that merely flag anomalous samples without quantifying the degree of anomaly. Moreover, a new dataset is gathered from an actuator vibrating at specific frequencies to benchmark the algorithms and evaluate the framework. Novel conditions are introduced by altering the input wave signal. Our findings offer valuable insights into the adaptability and robustness of unsupervised learning techniques for real-world novelty detection applications.
Title:
Watchlist Challenge: 3rd Open-set Face Detection and Identification
Authors: Furkan Kasım, Terrance E. Boult, Rensso Mora, Bernardo Biesseck, Rafael Ribeiro, Jan Schlueter, Tomáš Repák, Rafael Henrique Vareto, David Menotti, William Robson Schwartz, Manuel Günther
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
In the current landscape of biometrics and surveillance, the ability to accurately recognize faces in uncontrolled settings is paramount. The Watchlist Challenge addresses this critical need by focusing on face detection and open-set identification in real-world surveillance scenarios. This paper presents a comprehensive evaluation of participating algorithms, using the enhanced UnConstrained College Students (UCCS) dataset with new evaluation protocols. In total, four participants submitted four face detection and nine open-set face recognition systems. The evaluation demonstrates that while detection capabilities are generally robust, closed-set identification performance varies significantly, with models pre-trained on large-scale datasets showing superior performance. However, open-set scenarios require further improvement, especially at higher true positive identification rates, i.e., lower thresholds.
Title:
Muskits-ESPnet: A Comprehensive Toolkit for Singing Voice Synthesis in New Paradigm
Abstract
This research presents Muskits-ESPnet, a versatile toolkit that introduces new paradigms to Singing Voice Synthesis (SVS) through the application of pretrained audio models in both continuous and discrete approaches. Specifically, we explore discrete representations derived from SSL models and audio codecs and offer significant advantages in versatility and intelligence, supporting multi-format inputs and adaptable data processing workflows for various SVS models. The toolkit features automatic music score error detection and correction, as well as a perception auto-evaluation module to imitate human subjective evaluating scores. Muskits-ESPnet is available at \url{this https URL}.
Title:
Diff-VPS: Video Polyp Segmentation via a Multi-task Diffusion Network with Adversarial Temporal Reasoning
Abstract
Diffusion Probabilistic Models have recently attracted significant attention in the community of computer vision due to their outstanding performance. However, while a substantial amount of diffusion-based research has focused on generative tasks, no work introduces diffusion models to advance the results of polyp segmentation in videos, which is frequently challenged by polyps' high camouflage and redundant temporal this http URL this paper, we present a novel diffusion-based network for video polyp segmentation task, dubbed as Diff-VPS. We incorporate multi-task supervision into diffusion models to promote the discrimination of diffusion models on pixel-by-pixel segmentation. This integrates the contextual high-level information achieved by the joint classification and detection tasks. To explore the temporal dependency, Temporal Reasoning Module (TRM) is devised via reasoning and reconstructing the target frame from the previous frames. We further equip TRM with a generative adversarial self-supervised strategy to produce more realistic frames and thus capture better dynamic cues. Extensive experiments are conducted on SUN-SEG, and the results indicate that our proposed Diff-VPS significantly achieves state-of-the-art performance. Code is available at this https URL.
Title:
TLD-READY: Traffic Light Detection -- Relevance Estimation and Deployment Analysis
Authors: Nikolai Polley, Svetlana Pavlitska, Yacin Boualili, Patrick Rohrbeck, Paul Stiller, Ashok Kumar Bangaru, J. Marius Zöllner
Abstract
Effective traffic light detection is a critical component of the perception stack in autonomous vehicles. This work introduces a novel deep-learning detection system while addressing the challenges of previous work. Utilizing a comprehensive dataset amalgamation, including the Bosch Small Traffic Lights Dataset, LISA, the DriveU Traffic Light Dataset, and a proprietary dataset from Karlsruhe, we ensure a robust evaluation across varied scenarios. Furthermore, we propose a relevance estimation system that innovatively uses directional arrow markings on the road, eliminating the need for prior map creation. On the DriveU dataset, this approach results in 96% accuracy in relevance estimation. Finally, a real-world evaluation is performed to evaluate the deployment and generalizing abilities of these models. For reproducibility and to facilitate further research, we provide the model weights and code: this https URL.
Title:
MEDIC: Towards a Comprehensive Framework for Evaluating LLMs in Clinical Applications
Authors: Praveen K Kanithi, Clément Christophe, Marco AF Pimentel, Tathagata Raha, Nada Saadi, Hamza Javed, Svetlana Maslenkova, Nasir Hayat, Ronnie Rajan, Shadab Khan
Subjects: Subjects:
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Abstract
The rapid development of Large Language Models (LLMs) for healthcare applications has spurred calls for holistic evaluation beyond frequently-cited benchmarks like USMLE, to better reflect real-world performance. While real-world assessments are valuable indicators of utility, they often lag behind the pace of LLM evolution, likely rendering findings obsolete upon deployment. This temporal disconnect necessitates a comprehensive upfront evaluation that can guide model selection for specific clinical applications. We introduce MEDIC, a framework assessing LLMs across five critical dimensions of clinical competence: medical reasoning, ethics and bias, data and language understanding, in-context learning, and clinical safety. MEDIC features a novel cross-examination framework quantifying LLM performance across areas like coverage and hallucination detection, without requiring reference outputs. We apply MEDIC to evaluate LLMs on medical question-answering, safety, summarization, note generation, and other tasks. Our results show performance disparities across model sizes, baseline vs medically finetuned models, and have implications on model selection for applications requiring specific model strengths, such as low hallucination or lower cost of inference. MEDIC's multifaceted evaluation reveals these performance trade-offs, bridging the gap between theoretical capabilities and practical implementation in healthcare settings, ensuring that the most promising models are identified and adapted for diverse healthcare applications.
Title:
GitSEED: A Git-backed Automated Assessment Tool for Software Engineering and Programming Education
Authors: Pedro Orvalho, Mikoláš Janota, Vasco Manquinho
Subjects: Subjects:
Software Engineering (cs.SE); Computers and Society (cs.CY)
Abstract
Due to the substantial number of enrollments in programming courses, a key challenge is delivering personalized feedback to students. The nature of this feedback varies significantly, contingent on the subject and the chosen evaluation method. However, tailoring current Automated Assessment Tools (AATs) to integrate other program analysis tools is not straightforward. Moreover, AATs usually support only specific programming languages, providing feedback exclusively through dedicated websites based on test suites. This paper introduces GitSEED, a language-agnostic automated assessment tool designed for Programming Education and Software Engineering (SE) and backed by GitLab. The students interact with GitSEED through GitLab. Using GitSEED, students in Computer Science (CS) and SE can master the fundamentals of git while receiving personalized feedback on their programming assignments and projects. Furthermore, faculty members can easily tailor GitSEED's pipeline by integrating various code evaluation tools (e.g., memory leak detection, fault localization, program repair, etc.) to offer personalized feedback that aligns with the needs of each CS/SE course. Our experiments assess GitSEED's efficacy via comprehensive user evaluation, examining the impact of feedback mechanisms and features on student learning outcomes. Findings reveal positive correlations between GitSEED usage and student engagement.
Abstract
The increasing reliance on machine learning (ML) in computer security, particularly for malware classification, has driven significant advancements. However, the replicability and reproducibility of these results are often overlooked, leading to challenges in verifying research findings. This paper highlights critical pitfalls that undermine the validity of ML research in Android malware detection, focusing on dataset and methodological issues. We comprehensively analyze Android malware detection using two datasets and assess offline and continual learning settings with six widely used ML models. Our study reveals that when properly tuned, simpler baseline methods can often outperform more complex models. To address reproducibility challenges, we propose solutions for improving datasets and methodological practices, enabling fairer model comparisons. Additionally, we open-source our code to facilitate malware analysis, making it extensible for new models and datasets. Our paper aims to support future research in Android malware detection and other security domains, enhancing the reliability and reproducibility of published results.
Title:
Towards Fairer Health Recommendations: finding informative unbiased samples via Word Sense Disambiguation
Abstract
There have been growing concerns around high-stake applications that rely on models trained with biased data, which consequently produce biased predictions, often harming the most vulnerable. In particular, biased medical data could cause health-related applications and recommender systems to create outputs that jeopardize patient care and widen disparities in health outcomes. A recent framework titled Fairness via AI posits that, instead of attempting to correct model biases, researchers must focus on their root causes by using AI to debias data. Inspired by this framework, we tackle bias detection in medical curricula using NLP models, including LLMs, and evaluate them on a gold standard dataset containing 4,105 excerpts annotated by medical experts for bias from a large corpus. We build on previous work by coauthors which augments the set of negative samples with non-annotated text containing social identifier terms. However, some of these terms, especially those related to race and ethnicity, can carry different meanings (e.g., "white matter of spinal cord"). To address this issue, we propose the use of Word Sense Disambiguation models to refine dataset quality by removing irrelevant sentences. We then evaluate fine-tuned variations of BERT models as well as GPT models with zero- and few-shot prompting. We found LLMs, considered SOTA on many NLP tasks, unsuitable for bias detection, while fine-tuned BERT models generally perform well across all evaluated metrics.
Title:
Introducing Perturb-ability Score (PS) to Enhance Robustness Against Evasion Adversarial Attacks on ML-NIDS
Abstract
This paper proposes a novel Perturb-ability Score (PS) that can be used to identify Network Intrusion Detection Systems (NIDS) features that can be easily manipulated by attackers in the problem-space. We demonstrate that using PS to select only non-perturb-able features for ML-based NIDS maintains detection performance while enhancing robustness against adversarial attacks.
Keyword: face recognition
Title:
Face Mask Removal with Region-attentive Face Inpainting
Authors: Minmin Yang
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
During the COVID-19 pandemic, face masks have become ubiquitous in our lives. Face masks can cause some face recognition models to fail since they cover significant portion of a face. In addition, removing face masks from captured images or videos can be desirable, e.g., for better social interaction and for image/video editing and enhancement purposes. Hence, we propose a generative face inpainting method to effectively recover/reconstruct the masked part of a face. Face inpainting is more challenging compared to traditional inpainting, since it requires high fidelity while maintaining the identity at the same time. Our proposed method includes a Multi-scale Channel-Spatial Attention Module (M-CSAM) to mitigate the spatial information loss and learn the inter- and intra-channel correlation. In addition, we introduce an approach enforcing the supervised signal to focus on masked regions instead of the whole image. We also synthesize our own Masked-Faces dataset from the CelebA dataset by incorporating five different types of face masks, including surgical mask, regular mask and scarves, which also cover the neck area. The experimental results show that our proposed method outperforms different baselines in terms of structural similarity index measure, peak signal-to-noise ratio and l1 loss, while also providing better outputs qualitatively. The code will be made publicly available. Code is available at GitHub.
Title:
Watchlist Challenge: 3rd Open-set Face Detection and Identification
Authors: Furkan Kasım, Terrance E. Boult, Rensso Mora, Bernardo Biesseck, Rafael Ribeiro, Jan Schlueter, Tomáš Repák, Rafael Henrique Vareto, David Menotti, William Robson Schwartz, Manuel Günther
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
In the current landscape of biometrics and surveillance, the ability to accurately recognize faces in uncontrolled settings is paramount. The Watchlist Challenge addresses this critical need by focusing on face detection and open-set identification in real-world surveillance scenarios. This paper presents a comprehensive evaluation of participating algorithms, using the enhanced UnConstrained College Students (UCCS) dataset with new evaluation protocols. In total, four participants submitted four face detection and nine open-set face recognition systems. The evaluation demonstrates that while detection capabilities are generally robust, closed-set identification performance varies significantly, with models pre-trained on large-scale datasets showing superior performance. However, open-set scenarios require further improvement, especially at higher true positive identification rates, i.e., lower thresholds.
Keyword: augmentation
Title:
Feedback-based Modal Mutual Search for Attacking Vision-Language Pre-training Models
Authors: Renhua Ding, Xinze Zhang, Xiao Yang, Kun He
Abstract
Although vision-language pre-training (VLP) models have achieved remarkable progress on cross-modal tasks, they remain vulnerable to adversarial attacks. Using data augmentation and cross-modal interactions to generate transferable adversarial examples on surrogate models, transfer-based black-box attacks have become the mainstream methods in attacking VLP models, as they are more practical in real-world scenarios. However, their transferability may be limited due to the differences on feature representation across different models. To this end, we propose a new attack paradigm called Feedback-based Modal Mutual Search (FMMS). FMMS introduces a novel modal mutual loss (MML), aiming to push away the matched image-text pairs while randomly drawing mismatched pairs closer in feature space, guiding the update directions of the adversarial examples. Additionally, FMMS leverages the target model feedback to iteratively refine adversarial examples, driving them into the adversarial region. To our knowledge, this is the first work to exploit target model feedback to explore multi-modality adversarial boundaries. Extensive empirical evaluations on Flickr30K and MSCOCO datasets for image-text matching tasks show that FMMS significantly outperforms the state-of-the-art baselines.
Title:
The Competition Complexity of Prophet Inequalities with Correlations
Authors: Tomer Ezra, Tamar Garbuz
Subjects: Subjects:
Machine Learning (cs.LG); Computer Science and Game Theory (cs.GT)
Abstract
We initiate the study of the prophet inequality problem through the resource augmentation framework in scenarios when the values of the rewards are correlated. Our goal is to determine the number of additional rewards an online algorithm requires to approximate the maximum value of the original instance. While the independent reward case is well understood, we extend this research to account for correlations among rewards. Our results demonstrate that, unlike in the independent case, the required number of additional rewards for approximation depends on the number of original rewards, and that block-threshold algorithms, which are optimal in the independent case, may require an infinite number of additional rewards when correlations are present. We develop asymptotically optimal algorithms for the following three scenarios: (1) where rewards arrive in blocks corresponding to the different copies of the original instance; (2) where rewards across all copies are arbitrarily shuffled; and (3) where rewards arrive in blocks corresponding to the different copies of the original instance, and values within each block are pairwise independent rather than fully correlated.
Title:
Bridging Domain Gap of Point Cloud Representations via Self-Supervised Geometric Augmentation
Authors: Li Yu, Hongchao Zhong, Longkun Zou, Ke Chen, Pan Gao
Subjects: Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Abstract
Recent progress of semantic point clouds analysis is largely driven by synthetic data (e.g., the ModelNet and the ShapeNet), which are typically complete, well-aligned and noisy free. Therefore, representations of those ideal synthetic point clouds have limited variations in the geometric perspective and can gain good performance on a number of 3D vision tasks such as point cloud classification. In the context of unsupervised domain adaptation (UDA), representation learning designed for synthetic point clouds can hardly capture domain invariant geometric patterns from incomplete and noisy point clouds. To address such a problem, we introduce a novel scheme for induced geometric invariance of point cloud representations across domains, via regularizing representation learning with two self-supervised geometric augmentation tasks. On one hand, a novel pretext task of predicting translation distances of augmented samples is proposed to alleviate centroid shift of point clouds due to occlusion and noises. On the other hand, we pioneer an integration of the relational self-supervised learning on geometrically-augmented point clouds in a cascade manner, utilizing the intrinsic relationship of augmented variants and other samples as extra constraints of cross-domain geometric features. Experiments on the PointDA-10 dataset demonstrate the effectiveness of the proposed method, achieving the state-of-the-art performance.
Title:
Improving Anomalous Sound Detection via Low-Rank Adaptation Fine-Tuning of Pre-Trained Audio Models
Abstract
Anomalous Sound Detection (ASD) has gained significant interest through the application of various Artificial Intelligence (AI) technologies in industrial settings. Though possessing great potential, ASD systems can hardly be readily deployed in real production sites due to the generalization problem, which is primarily caused by the difficulty of data collection and the complexity of environmental factors. This paper introduces a robust ASD model that leverages audio pre-trained models. Specifically, we fine-tune these models using machine operation data, employing SpecAug as a data augmentation strategy. Additionally, we investigate the impact of utilizing Low-Rank Adaptation (LoRA) tuning instead of full fine-tuning to address the problem of limited data for fine-tuning. Our experiments on the DCASE2023 Task 2 dataset establish a new benchmark of 77.75% on the evaluation set, with a significant improvement of 6.48% compared with previous state-of-the-art (SOTA) models, including top-tier traditional convolutional networks and speech pre-trained models, which demonstrates the effectiveness of audio pre-trained models with LoRA tuning. Ablation studies are also conducted to showcase the efficacy of the proposed scheme.
Title:
Data Augmentation via Latent Diffusion for Saliency Prediction
Abstract
Saliency prediction models are constrained by the limited diversity and quantity of labeled data. Standard data augmentation techniques such as rotating and cropping alter scene composition, affecting saliency. We propose a novel data augmentation method for deep saliency prediction that edits natural images while preserving the complexity and variability of real-world scenes. Since saliency depends on high-level and low-level features, our approach involves learning both by incorporating photometric and semantic attributes such as color, contrast, brightness, and class. To that end, we introduce a saliency-guided cross-attention mechanism that enables targeted edits on the photometric properties, thereby enhancing saliency within specific image regions. Experimental results show that our data augmentation method consistently improves the performance of various saliency models. Moreover, leveraging the augmentation features for saliency prediction yields superior performance on publicly available saliency benchmarks. Our predictions align closely with human visual attention patterns in the edited images, as validated by a user study.
Abstract
Pretraining on large-scale, unstructured internet text has enabled language models to acquire a significant amount of world knowledge. However, this knowledge acquisition is data-inefficient -- to learn a given fact, models must be trained on hundreds to thousands of diverse representations of it. This poses a challenge when adapting a pretrained model to a small corpus of domain-specific documents, where each fact may appear rarely or only once. We propose to bridge this gap with synthetic continued pretraining: using the small domain-specific corpus to synthesize a large corpus more amenable to learning, and then performing continued pretraining on the synthesized corpus. We instantiate this proposal with EntiGraph, a synthetic data augmentation algorithm that extracts salient entities from the source documents and then generates diverse text by drawing connections between the sampled entities. Synthetic continued pretraining using EntiGraph enables a language model to answer questions and follow generic instructions related to the source documents without access to them. If instead, the source documents are available at inference time, we show that the knowledge acquired through our approach compounds with retrieval-augmented generation. To better understand these results, we build a simple mathematical model of EntiGraph, and show how synthetic data augmentation can "rearrange" knowledge to enable more data-efficient learning.
Keyword: detection
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Title:
Keyword: face recognition
Title:
Title:
Keyword: augmentation
Title:
Title:
Title:
Title:
Title:
Title: