Search

We establish the existence of non-stationary solutions to a symmetric system of second-order autonomous differential equations. Our technique is based on the equivariant degree theory and involves a novel characterization of orbit types of maximal kind in the Burnside Ring product of a finite number of basic degrees for the group $O(2) \times \Gamma \times \mathbb Z_2$.

This Research proposes a Novel Reinforcement Learning (RL) model to optimise malware forensics investigation during cyber incident response. It aims to improve forensic investigation efficiency by reducing false negatives and adapting current practices to evolving malware signatures. The proposed RL framework leverages techniques such as Q-learning and the Markov Decision Process (MDP) to train the system to identify malware patterns in live memory dumps, thereby automating forensic tasks. The RL model is based on a detailed malware workflow diagram that guides the analysis of malware artefacts using static and behavioural techniques as well as machine learning algorithms. Furthermore, it seeks to address challenges in the UK justice system by ensuring the accuracy of forensic evidence. We conduct testing and evaluation in controlled environments, using datasets created with Windows operating systems to simulate malware infections. The experimental results demonstrate that RL improves malware detection rates compared to conventional methods, with the RL model's performance varying depending on the complexity and learning rate of the environment. The study concludes that while RL offers promising potential for automating malware forensics, its efficacy across diverse malware types requires ongoing refinement of reward systems and feature extraction methods., Comment: 8 pages. arXiv admin note: substantial text overlap with arXiv:2408.01999

This work presents a novel framework for physically consistent model error characterization and operator learning for reduced-order models of non-equilibrium chemical kinetics. By leveraging the Bayesian framework, we identify and infer sources of model and parametric uncertainty within the Coarse-Graining Methodology across a range of initial conditions. The model error is embedded into the chemical kinetics model to ensure that its propagation to quantities of interest remains physically consistent. For operator learning, we develop a methodology that separates time dynamics from other input parameters. Karhunen-Loeve Expansion (KLE) is employed to capture time dynamics, yielding temporal modes, while Polynomial Chaos Expansion (PCE) is subsequently used to map model error and input parameters to KLE coefficients. The proposed model offers three significant advantages: i) Separating time dynamics from other inputs ensures stability of chemistry surrogate when coupled with fluid solvers; ii) The framework fully accounts for model and parametric uncertainty, enabling robust probabilistic predictions; iii) The surrogate model is highly interpretable, with visualizable time modes and a PCE component that facilitates analytical calculation of sensitivity indices. We apply this framework to O2-O chemistry system under hypersonic flight conditions, validating it in both a 0D adiabatic reactor and coupled simulations with a fluid solver in a 1D shock case. Results demonstrate that the surrogate is stable during time integration, delivers physically consistent probabilistic predictions accounting for model and parametric uncertainty, and achieves maximum relative error below 10%. This work represents a significant step forward in enabling probabilistic predictions of non-equilibrium chemistry with coupled fluid solvers, offering a physically accurate approach for hypersonic flow predictions.

This paper investigates the application of Deep Reinforcement Learning (DRL) for attributing malware to specific Advanced Persistent Threat (APT) groups through detailed behavioural analysis. By analysing over 3500 malware samples from 12 distinct APT groups, the study utilises sophisticated tools like Cuckoo Sandbox to extract behavioural data, providing a deep insight into the operational patterns of malware. The research demonstrates that the DRL model significantly outperforms traditional machine learning approaches such as SGD, SVC, KNN, MLP, and Decision Tree Classifiers, achieving an impressive test accuracy of 89.27 %. It highlights the model capability to adeptly manage complex, variable, and elusive malware attributes. Furthermore, the paper discusses the considerable computational resources and extensive data dependencies required for deploying these advanced AI models in cybersecurity frameworks. Future research is directed towards enhancing the efficiency of DRL models, expanding the diversity of the datasets, addressing ethical concerns, and leveraging Large Language Models (LLMs) to refine reward mechanisms and optimise the DRL framework. By showcasing the transformative potential of DRL in malware attribution, this research advocates for a responsible and balanced approach to AI integration, with the goal of advancing cybersecurity through more adaptable, accurate, and robust systems., Comment: 21 Pages

The task of action spotting consists in both identifying actions and precisely localizing them in time with a single timestamp in long, untrimmed video streams. Automatically extracting those actions is crucial for many sports applications, including sports analytics to produce extended statistics on game actions, coaching to provide support to video analysts, or fan engagement to automatically overlay content in the broadcast when specific actions occur. However, before 2018, no large-scale datasets for action spotting in sports were publicly available, which impeded benchmarking action spotting methods. In response, our team built the largest dataset and the most comprehensive benchmarks for sports video understanding, under the umbrella of SoccerNet. Particularly, our dataset contains a subset specifically dedicated to action spotting, called SoccerNet Action Spotting, containing more than 550 complete broadcast games annotated with almost all types of actions that can occur in a football game. This dataset is tailored to develop methods for automatic spotting of actions of interest, including deep learning approaches, by providing a large amount of manually annotated actions. To engage with the scientific community, the SoccerNet initiative organizes yearly challenges, during which participants from all around the world compete to achieve state-of-the-art performances. Thanks to our dataset and challenges, more than 60 methods were developed or published over the past five years, improving on the first baselines and making action spotting a viable option for the sports industry. This paper traces the history of action spotting in sports, from the creation of the task back in 2018, to the role it plays today in research and the sports industry., Comment: 31 pages, 2 figures, 5 tables

Accurate assessment of disease severity from endoscopy videos in ulcerative colitis (UC) is crucial for evaluating drug efficacy in clinical trials. Severity is often measured by the Mayo Endoscopic Subscore (MES) and Ulcerative Colitis Endoscopic Index of Severity (UCEIS) score. However, expert MES/UCEIS annotation is time-consuming and susceptible to inter-rater variability, factors addressable by automation. Automation attempts with frame-level labels face challenges in fully-supervised solutions due to the prevalence of video-level labels in clinical trials. CNN-based weakly-supervised models (WSL) with end-to-end (e2e) training lack generalization to new disease scores and ignore spatio-temporal information crucial for accurate scoring. To address these limitations, we propose "Arges", a deep learning framework that utilizes a transformer with positional encoding to incorporate spatio-temporal information from frame features to estimate disease severity scores in endoscopy video. Extracted features are derived from a foundation model (ArgesFM), pre-trained on a large diverse dataset from multiple clinical trials (61M frames, 3927 videos). We evaluate four UC disease severity scores, including MES and three UCEIS component scores. Test set evaluation indicates significant improvements, with F1 scores increasing by 4.1% for MES and 18.8%, 6.6%, 3.8% for the three UCEIS component scores compared to state-of-the-art methods. Prospective validation on previously unseen clinical trial data further demonstrates the model's successful generalization., Comment: 12 pages, 2 figures, 5 tables, accepted at MLMI, MICCAI

The SoccerNet 2024 challenges represent the fourth annual video understanding challenges organized by the SoccerNet team. These challenges aim to advance research across multiple themes in football, including broadcast video understanding, field understanding, and player understanding. This year, the challenges encompass four vision-based tasks. (1) Ball Action Spotting, focusing on precisely localizing when and which soccer actions related to the ball occur, (2) Dense Video Captioning, focusing on describing the broadcast with natural language and anchored timestamps, (3) Multi-View Foul Recognition, a novel task focusing on analyzing multiple viewpoints of a potential foul incident to classify whether a foul occurred and assess its severity, (4) Game State Reconstruction, another novel task focusing on reconstructing the game state from broadcast videos onto a 2D top-view map of the field. Detailed information about the tasks, challenges, and leaderboards can be found at https://www.soccer-net.org, with baselines and development kits available at https://github.com/SoccerNet., Comment: 7 pages, 1 figure

The utility of solving the Fermi-Hubbard model has been estimated in the billions of dollars. Digital quantum computers can in principle address this task, but have so far been limited to quasi one-dimensional models. This is because of exponential overheads caused by the interplay of noise and the non-locality of the mapping between fermions and qubits. Here, we show experimentally that a recently developed local encoding can overcome this problem. We develop a new compilation scheme, called "corner hopping", that reduces the cost of simulating fermionic hopping by 42% which allows us to conduct the largest digital quantum simulations of a fermionic model to date, using a trapped ion quantum computer to prepare adiabatically the ground state of a 6 x 6 spinless Fermi-Hubbard model encoded in 48 physical qubits. We also develop two new error mitigation schemes for systems with conserved quantities, one based on local postselection and one on extrapolation of local observables. Our results suggest that Fermi-Hubbard models beyond classical simulability can be addressed by digital quantum computers without large increases in gate fidelity., Comment: 26 pages, 21 figures

Timeline Analysis (TA) plays a crucial role in Timeline Forensics (TF) within the field of Digital Forensics (DF). It focuses on examining and analyzing time-based digital artefacts, such as timestamps derived from event logs, file metadata, and other relevant data, to correlate events linked to cyber incidents and reconstruct their chronological sequence. Traditional tools often struggle to efficiently handle the large volume and variety of data generated during DF investigations and Incident Response (IR) processes. This paper introduces a novel framework, GenDFIR, which combines Rule-Based Artificial Intelligence (R-BAI) algorithms with Large Language Models (LLMs) to enhance and automate the TA process. The proposed approach consists of two key stages: (1) R-BAI is used to identify and select anomalous digital artefacts based on predefined rules. (2) The selected artefacts are then transformed into embeddings for processing by an LLM with the assistance of a Retrieval-Augmented Generation (RAG) agent. The LLM uses its capabilities to perform automated TA on the artefacts and predict potential incident outcomes. To validate the framework, we evaluated its performance, efficiency, and reliability. Several metrics were applied to simulated cyber incident scenarios, which were presented as forensic case documents. Our findings demonstrate the significant potential of integrating R-BAI and LLMs for TA. This innovative approach underscores the power of Generative AI (GenAI), particularly LLMs, and opens up new possibilities for advanced threat detection and incident reconstruction, marking a significant advancement in the field., Comment: 22 pages V3.1

Given the rapidly expanding capabilities of generative AI models for radiology, there is a need for robust metrics that can accurately measure the quality of AI-generated radiology reports across diverse hospitals. We develop ReXamine-Global, a LLM-powered, multi-site framework that tests metrics across different writing styles and patient populations, exposing gaps in their generalization. First, our method tests whether a metric is undesirably sensitive to reporting style, providing different scores depending on whether AI-generated reports are stylistically similar to ground-truth reports or not. Second, our method measures whether a metric reliably agrees with experts, or whether metric and expert scores of AI-generated report quality diverge for some sites. Using 240 reports from 6 hospitals around the world, we apply ReXamine-Global to 7 established report evaluation metrics and uncover serious gaps in their generalizability. Developers can apply ReXamine-Global when designing new report evaluation metrics, ensuring their robustness across sites. Additionally, our analysis of existing metrics can guide users of those metrics towards evaluation procedures that work reliably at their sites of interest.

This paper presents preliminary work on a novel connection between certified robustness in machine learning and the modeling of 3D objects. We highlight an intriguing link between the Maximal Certified Radius (MCR) of a classifier representing a space's occupancy and the space's Signed Distance Function (SDF). Leveraging this relationship, we propose to use the certification method of randomized smoothing (RS) to compute SDFs. Since RS' high computational cost prevents its practical usage as a way to compute SDFs, we propose an algorithm to efficiently run RS in low-dimensional applications, such as 3D space, by expressing RS' fundamental operations as Gaussian smoothing on pre-computed voxel grids. Our approach offers an innovative and practical tool to compute SDFs, validated through proof-of-concept experiments in novel view synthesis. This paper bridges two previously disparate areas of machine learning, opening new avenues for further exploration and potential cross-domain advancements., Comment: This paper is an accepted extended abstract to the LatinX workshop at ICCV 2023. This was uploaded a year late

Chronic wounds pose an ongoing health concern globally, largely due to the prevalence of conditions such as diabetes and leprosy's disease. The standard method of monitoring these wounds involves visual inspection by healthcare professionals, a practice that could present challenges for patients in remote areas with inadequate transportation and healthcare infrastructure. This has led to the development of algorithms designed for the analysis and follow-up of wound images, which perform image-processing tasks such as classification, detection, and segmentation. However, the effectiveness of these algorithms heavily depends on the availability of comprehensive and varied wound image data, which is usually scarce. This paper introduces the CO2Wounds-V2 dataset, an extended collection of RGB wound images from leprosy patients with their corresponding semantic segmentation annotations, aiming to enhance the development and testing of image-processing algorithms in the medical field., Comment: 2024 IEEE International Conference on Image Processing (ICIP 2024)

Neural radiance fields (NeRFs) generally require many images with accurate poses for accurate novel view synthesis, which does not reflect realistic setups where views can be sparse and poses can be noisy. Previous solutions for learning NeRFs with sparse views and noisy poses only consider local geometry consistency with pairs of views. Closely following \textit{bundle adjustment} in Structure-from-Motion (SfM), we introduce TrackNeRF for more globally consistent geometry reconstruction and more accurate pose optimization. TrackNeRF introduces \textit{feature tracks}, \ie connected pixel trajectories across \textit{all} visible views that correspond to the \textit{same} 3D points. By enforcing reprojection consistency among feature tracks, TrackNeRF encourages holistic 3D consistency explicitly. Through extensive experiments, TrackNeRF sets a new benchmark in noisy and sparse view reconstruction. In particular, TrackNeRF shows significant improvements over the state-of-the-art BARF and SPARF by $\sim8$ and $\sim1$ in terms of PSNR on DTU under various sparse and noisy view setups. The code is available at \href{https://tracknerf.github.io/}., Comment: ECCV 2024 (supplemental pages included)

Large Language Models (LLMs) offer significant potential for clinical symptom extraction, but their deployment in healthcare settings is constrained by privacy concerns, computational limitations, and operational costs. This study investigates the optimization of compact LLMs for cancer toxicity symptom extraction using a novel iterative refinement approach. We employ a student-teacher architecture, utilizing Zephyr-7b-beta and Phi3-mini-128 as student models and GPT-4o as the teacher, to dynamically select between prompt refinement, Retrieval-Augmented Generation (RAG), and fine-tuning strategies. Our experiments on 294 clinical notes covering 12 post-radiotherapy toxicity symptoms demonstrate the effectiveness of this approach. The RAG method proved most efficient, improving average accuracy scores from 0.32 to 0.73 for Zephyr-7b-beta and from 0.40 to 0.87 for Phi3-mini-128 during refinement. In the test set, both models showed an approximate 0.20 increase in accuracy across symptoms. Notably, this improvement was achieved at a cost 45 times lower than GPT-4o for Zephyr and 79 times lower for Phi-3. These results highlight the potential of iterative refinement techniques in enhancing the capabilities of compact LLMs for clinical applications, offering a balance between performance, cost-effectiveness, and privacy preservation in healthcare settings.

This research focused on enhancing post-incident malware forensic investigation using reinforcement learning RL. We proposed an advanced MDP post incident malware forensics investigation model and framework to expedite post incident forensics. We then implement our RL Malware Investigation Model based on structured MDP within the proposed framework. To identify malware artefacts, the RL agent acquires and examines forensics evidence files, iteratively improving its capabilities using Q Table and temporal difference learning. The Q learning algorithm significantly improved the agent ability to identify malware. An epsilon greedy exploration strategy and Q learning updates enabled efficient learning and decision making. Our experimental testing revealed that optimal learning rates depend on the MDP environment complexity, with simpler environments benefiting from higher rates for quicker convergence and complex ones requiring lower rates for stability. Our model performance in identifying and classifying malware reduced malware analysis time compared to human experts, demonstrating robustness and adaptability. The study highlighted the significance of hyper parameter tuning and suggested adaptive strategies for complex environments. Our RL based approach produced promising results and is validated as an alternative to traditional methods notably by offering continuous learning and adaptation to new and evolving malware threats which ultimately enhance the post incident forensics investigations., Comment: v1.1

PLoM (Probabilistic Learning on Manifolds) is a method introduced in 2016 for handling small training datasets by projecting an It\^o equation from a stochastic dissipative Hamiltonian dynamical system, acting as the MCMC generator, for which the KDE-estimated probability measure with the training dataset is the invariant measure. PLoM performs a projection on a reduced-order vector basis related to the training dataset, using the diffusion maps (DMAPS) basis constructed with a time-independent isotropic kernel. In this paper, we propose a new ISDE projection vector basis built from a transient anisotropic kernel, providing an alternative to the DMAPS basis to improve statistical surrogates for stochastic manifolds with heterogeneous data. The construction ensures that for times near the initial time, the DMAPS basis coincides with the transient basis. For larger times, the differences between the two bases are characterized by the angle of their spanned vector subspaces. The optimal instant yielding the optimal transient basis is determined using an estimation of mutual information from Information Theory, which is normalized by the entropy estimation to account for the effects of the number of realizations used in the estimations. Consequently, this new vector basis better represents statistical dependencies in the learned probability measure for any dimension. Three applications with varying levels of statistical complexity and data heterogeneity validate the proposed theory, showing that the transient anisotropic kernel improves the learned probability measure., Comment: 44 pages, 14 figures

As a fundamental task in long-form video understanding, temporal action detection (TAD) aims to capture inherent temporal relations in untrimmed videos and identify candidate actions with precise boundaries. Over the years, various networks, including convolutions, graphs, and transformers, have been explored for effective temporal modeling for TAD. However, these modules typically treat past and future information equally, overlooking the crucial fact that changes in action boundaries are essentially causal events. Inspired by this insight, we propose leveraging the temporal causality of actions to enhance TAD representation by restricting the model's access to only past or future context. We introduce CausalTAD, which combines causal attention and causal Mamba to achieve state-of-the-art performance on multiple benchmarks. Notably, with CausalTAD, we ranked 1st in the Action Recognition, Action Detection, and Audio-Based Interaction Detection tracks at the EPIC-Kitchens Challenge 2024, as well as 1st in the Moment Queries track at the Ego4D Challenge 2024. Our code is available at https://github.com/sming256/OpenTAD/., Comment: 1st in Moment Queries track at the Ego4D Challenge 2024; 1st in Action Recognition, Action Detection, and Audio-Based Interaction Detection tracks at the EPIC-Kitchens Challenge 2024

Over the past decade, the technology used by referees in football has improved substantially, enhancing the fairness and accuracy of decisions. This progress has culminated in the implementation of the Video Assistant Referee (VAR), an innovation that enables backstage referees to review incidents on the pitch from multiple points of view. However, the VAR is currently limited to professional leagues due to its expensive infrastructure and the lack of referees worldwide. In this paper, we present the semi-automated Video Assistant Referee System (VARS) that leverages the latest findings in multi-view video analysis. VARS sets a new state-of-the-art on the SoccerNet-MVFoul dataset, a multi-view video dataset of football fouls. Our VARS achieves a new state-of-the-art on the SoccerNet-MVFoul dataset by recognizing the type of foul in 50% of instances and the appropriate sanction in 46% of cases. Finally, we conducted a comparative study to investigate human performance in classifying fouls and their corresponding severity and compared these findings to our VARS. The results of our study highlight the potential of our VARS to reach human performance and support football refereeing across all levels of professional and amateur federations., Comment: The paper is subject to the peer review process of Sports Engineering

Masked AutoEncoders (MAE) have emerged as a robust self-supervised framework, offering remarkable performance across a wide range of downstream tasks. To increase the difficulty of the pretext task and learn richer visual representations, existing works have focused on replacing standard random masking with more sophisticated strategies, such as adversarial-guided and teacher-guided masking. However, these strategies depend on the input data thus commonly increasing the model complexity and requiring additional calculations to generate the mask patterns. This raises the question: Can we enhance MAE performance beyond random masking without relying on input data or incurring additional computational costs? In this work, we introduce a simple yet effective data-independent method, termed ColorMAE, which generates different binary mask patterns by filtering random noise. Drawing inspiration from color noise in image processing, we explore four types of filters to yield mask patterns with different spatial and semantic priors. ColorMAE requires no additional learnable parameters or computational overhead in the network, yet it significantly enhances the learned representations. We provide a comprehensive empirical evaluation, demonstrating our strategy's superiority in downstream tasks compared to random masking. Notably, we report an improvement of 2.72 in mIoU in semantic segmentation tasks relative to baseline MAE implementations., Comment: Work Accepted for Publication at ECCV 2024

Despite its simplicity and strong theoretical guarantees, adiabatic state preparation has received considerably less interest than variational approaches for the preparation of low-energy electronic structure states. Two major reasons for this are the large number of gates required for Trotterising time-dependent electronic structure Hamiltonians, as well as discretisation errors heating the state. We show that a recently proposed randomized algorithm, which implements exact adiabatic evolution without heating and with far fewer gates than Trotterisation, can overcome this problem. We develop three methods for measuring the energy of the prepared state in an efficient and noise-resilient manner, yielding chemically accurate results on a 4-qubit molecule in the presence of realistic gate noise, without the need for error mitigation. These findings suggest that adiabatic approaches to state preparation could play a key role in quantum chemistry simulations both in the era of noisy as well as error-corrected quantum computers., Comment: 17 pages

For medical imaging AI models to be clinically impactful, they must generalize. However, this goal is hindered by (i) diverse types of distribution shifts, such as temporal, demographic, and label shifts, and (ii) limited diversity in datasets that are siloed within single medical institutions. While these limitations have spurred interest in federated learning, current evaluation benchmarks fail to evaluate different shifts simultaneously. However, in real healthcare settings, multiple types of shifts co-exist, yet their impact on medical imaging performance remains unstudied. In response, we introduce FedMedICL, a unified framework and benchmark to holistically evaluate federated medical imaging challenges, simultaneously capturing label, demographic, and temporal distribution shifts. We comprehensively evaluate several popular methods on six diverse medical imaging datasets (totaling 550 GPU hours). Furthermore, we use FedMedICL to simulate COVID-19 propagation across hospitals and evaluate whether methods can adapt to pandemic changes in disease prevalence. We find that a simple batch balancing technique surpasses advanced methods in average performance across FedMedICL experiments. This finding questions the applicability of results from previous, narrow benchmarks in real-world medical settings., Comment: Accepted at MICCAI 2024. Code is available at: https://github.com/m1k2zoo/FedMedICL

We built our pipeline EgoLoc-v1, mainly inspired by EgoLoc. We propose a model ensemble strategy to improve the camera pose estimation part of the VQ3D task, which has been proven to be essential in previous work. The core idea is not only to do SfM for egocentric videos but also to do 2D-3D matching between existing 3D scans and 2D video frames. In this way, we have a hybrid SfM and camera relocalization pipeline, which can provide us with more camera poses, leading to higher QwP and overall success rate. Our method achieves the best performance regarding the most important metric, the overall success rate. We surpass previous state-of-the-art, the competitive EgoLoc, by $1.5\%$. The code is available at \url{https://github.com/Wayne-Mai/egoloc_v1}., Comment: 1st place winner of the 2024 Ego4D-Ego-Exo4D Challenge in VQ3D

Few-shot class-incremental learning (FSCIL) confronts the challenge of integrating new classes into a model with minimal training samples while preserving the knowledge of previously learned classes. Traditional methods widely adopt static adaptation relying on a fixed parameter space to learn from data that arrive sequentially, prone to overfitting to the current session. Existing dynamic strategies require the expansion of the parameter space continually, leading to increased complexity. In this study, we explore the potential of Selective State Space Models (SSMs) for FSCIL, leveraging its dynamic weights and strong ability in sequence modeling to address these challenges. Concretely, we propose a dual selective SSM projector that dynamically adjusts the projection parameters based on the intermediate features for dynamic adaptation. The dual design enables the model to maintain the robust features of base classes, while adaptively learning distinctive feature shifts for novel classes. Additionally, we develop a class-sensitive selective scan mechanism to guide dynamic adaptation. It minimizes the disruption to base-class representations caused by training on novel data, and meanwhile, forces the selective scan to perform in distinct patterns between base and novel classes. Experiments on miniImageNet, CUB-200, and CIFAR-100 demonstrate that our framework outperforms the existing state-of-the-art methods. The code is available at \url{https://github.com/xiaojieli0903/Mamba-FSCIL}., Comment: Code: https://github.com/xiaojieli0903/Mamba-FSCIL

Slow-motion replays provide a thrilling perspective on pivotal moments within sports games, offering a fresh and captivating visual experience. However, capturing slow-motion footage typically demands high-tech, expensive cameras and infrastructures. Deep learning Video Frame Interpolation (VFI) techniques have emerged as a promising avenue, capable of generating high-speed footage from regular camera feeds. Moreover, the utilization of event-based cameras has recently gathered attention as they provide valuable motion information between frames, further enhancing the VFI performances. In this work, we present a first investigation of event-based VFI models for generating sports slow-motion videos. Particularly, we design and implement a bi-camera recording setup, including an RGB and an event-based camera to capture sports videos, to temporally align and spatially register both cameras. Our experimental validation demonstrates that TimeLens, an off-the-shelf event-based VFI model, can effectively generate slow-motion footage for sports videos. This first investigation underscores the practical utility of event-based cameras in producing sports slow-motion content and lays the groundwork for future research endeavors in this domain.

Action spotting is crucial in sports analytics as it enables the precise identification and categorization of pivotal moments in sports matches, providing insights that are essential for performance analysis and tactical decision-making. The fragmentation of existing methodologies, however, impedes the progression of sports analytics, necessitating a unified codebase to support the development and deployment of action spotting for video analysis. In this work, we introduce OSL-ActionSpotting, a Python library that unifies different action spotting algorithms to streamline research and applications in sports video analytics. OSL-ActionSpotting encapsulates various state-of-the-art techniques into a singular, user-friendly framework, offering standardized processes for action spotting and analysis across multiple datasets. We successfully integrated three cornerstone action spotting methods into OSL-ActionSpotting, achieving performance metrics that match those of the original, disparate codebases. This unification within a single library preserves the effectiveness of each method and enhances usability and accessibility for researchers and practitioners in sports analytics. By bridging the gaps between various action spotting techniques, OSL-ActionSpotting significantly contributes to the field of sports video analysis, fostering enhanced analytical capabilities and collaborative research opportunities. The scalable and modularized design of the library ensures its long-term relevance and adaptability to future technological advancements in the domain.

The development of autonomous agents increasingly relies on Multimodal Language Models (MLMs) to perform tasks described in natural language with GUI environments, such as websites, desktop computers, or mobile phones. Existing benchmarks for MLM agents in interactive environments are limited by their focus on a single environment, lack of detailed and generalized evaluation methods, and the complexities of constructing tasks and evaluators. To overcome these limitations, we introduce Crab, the first agent benchmark framework designed to support cross-environment tasks, incorporating a graph-based fine-grained evaluation method and an efficient mechanism for task and evaluator construction. Our framework supports multiple devices and can be easily extended to any environment with a Python interface. Leveraging Crab, we developed a cross-platform Crab Benchmark-v0 comprising 120 tasks in computer desktop and mobile phone environments. We evaluated four advanced MLMs using different single and multi-agent system configurations on this benchmark. The experimental results demonstrate that the single agent with GPT-4o achieves the best completion ratio of 38.01%. All framework code, agent code, and task datasets are publicly available at https://github.com/camel-ai/crab.

Artificial spin ices (ASIs) arranged in square formations have been explored from the perspective of reconfigurable magnonics. A new frontier in ASIs is their three-dimensional (3D) extension. Here, we numerically explore the ferromagnetic resonance of square ASIs as each nanomagnet is rotated out of plane into 3D ASIs, in which the vertex gap can be either kept constant or varying. We study both remanent and vortex configurations using a semi-analytical dynamic approach and micromagnetic simulations. We find that both methods show qualitative agreement of the main spectral features. However, there are important limitations. On one hand, the semi-analytical approach relies on a minimal model of the demag field, preventing exact predictions. On the other hand, micromagnetic simulations suffer from sufficient resolution, making the results grid-dependent and extremely slow. Regardless, both methods display tunability as a function of the tilt angle. These results showcase advantages and limitations of both methods and are promising to further our understanding of 3D ASI dynamics.

Merging Large Language Models (LLMs) is a cost-effective technique for combining multiple expert LLMs into a single versatile model, retaining the expertise of the original ones. However, current approaches often overlook the importance of safety alignment during merging, leading to highly misaligned models. This work investigates the effects of model merging on alignment. We evaluate several popular model merging techniques, demonstrating that existing methods do not only transfer domain expertise but also propagate misalignment. We propose a simple two-step approach to address this problem: (i) generating synthetic safety and domain-specific data, and (ii) incorporating these generated data into the optimization process of existing data-aware model merging techniques. This allows us to treat alignment as a skill that can be maximized in the resulting merged LLM. Our experiments illustrate the effectiveness of integrating alignment-related data during merging, resulting in models that excel in both domain expertise and alignment., Comment: Under review

While diffusion models have shown impressive performance in 2D image/video generation, diffusion-based Text-to-Multi-view-Video (T2MVid) generation remains underexplored. The new challenges posed by T2MVid generation lie in the lack of massive captioned multi-view videos and the complexity of modeling such multi-dimensional distribution. To this end, we propose a novel diffusion-based pipeline that generates high-quality multi-view videos centered around a dynamic 3D object from text. Specifically, we factor the T2MVid problem into viewpoint-space and time components. Such factorization allows us to combine and reuse layers of advanced pre-trained multi-view image and 2D video diffusion models to ensure multi-view consistency as well as temporal coherence for the generated multi-view videos, largely reducing the training cost. We further introduce alignment modules to align the latent spaces of layers from the pre-trained multi-view and the 2D video diffusion models, addressing the reused layers' incompatibility that arises from the domain gap between 2D and multi-view data. In support of this and future research, we further contribute a captioned multi-view video dataset. Experimental results demonstrate that our method generates high-quality multi-view videos, exhibiting vivid motions, temporal coherence, and multi-view consistency, given a variety of text prompts., Comment: Our project page is at https://hi-zhengcheng.github.io/vividzoo/

Relieving the reliance of neural network training on a global back-propagation (BP) has emerged as a notable research topic due to the biological implausibility and huge memory consumption caused by BP. Among the existing solutions, local learning optimizes gradient-isolated modules of a neural network with local errors and has been proved to be effective even on large-scale datasets. However, the reconciliation among local errors has never been investigated. In this paper, we first theoretically study non-greedy layer-wise training and show that the convergence cannot be assured when the local gradient in a module w.r.t. its input is not reconciled with the local gradient in the previous module w.r.t. its output. Inspired by the theoretical result, we further propose a local training strategy that successively regularizes the gradient reconciliation between neighboring modules without breaking gradient isolation or introducing any learnable parameters. Our method can be integrated into both local-BP and BP-free settings. In experiments, we achieve significant performance improvements compared to previous methods. Particularly, our method for CNN and Transformer architectures on ImageNet is able to attain a competitive performance with global BP, saving more than 40% memory consumption., Comment: ICML 2024

Current parameter-efficient fine-tuning (PEFT) methods build adapters widely agnostic of the context of downstream task to learn, or the context of important knowledge to maintain. As a result, there is often a performance gap compared to full-parameter fine-tuning, and meanwhile the fine-tuned model suffers from catastrophic forgetting of the pre-trained world knowledge. In this paper, we propose CorDA, a Context-oriented Decomposition Adaptation method that builds learnable task-aware adapters from weight decomposition oriented by the context of downstream task or the world knowledge to maintain. Concretely, we collect a few data samples, and perform singular value decomposition for each linear layer of a pre-trained LLM multiplied by the covariance matrix of the input activation using these samples. The inverse of the covariance matrix is multiplied with the decomposed components to reconstruct the original weights. By doing so, the context of the representative samples is captured through deciding the factorizing orientation. Our method enables two options, the knowledge-preserved adaptation and the instruction-previewed adaptation. For the former, we use question-answering samples to obtain the covariance matrices, and use the decomposed components with the smallest $r$ singular values to initialize a learnable adapter, with the others frozen such that the world knowledge is better preserved. For the latter, we use the instruction data from the fine-tuning task, such as math or coding, to orientate the decomposition and train the largest $r$ components that most correspond to the task to learn. We conduct extensive experiments on Math, Code, and Instruction Following tasks., Comment: NeurIPS 2024

Continuously optimizing sensor placement is essential for precise target localization in various military and civilian applications. While information theory has shown promise in optimizing sensor placement, many studies oversimplify sensor measurement models or neglect dynamic constraints of mobile sensors. To address these challenges, we employ a range measurement model that incorporates radar parameters and radar-target distance, coupled with Model Predictive Path Integral (MPPI) control to manage complex environmental obstacles and dynamic constraints. We compare the proposed approach against stationary radars or simplified range measurement models based on the root mean squared error (RMSE) of the Cubature Kalman Filter (CKF) estimator for the targets' state. Additionally, we visualize the evolving geometry of radars and targets over time, highlighting areas of highest measurement information gain, demonstrating the strengths of the approach. The proposed strategy outperforms stationary radars and simplified range measurement models in target localization, achieving a 38-74% reduction in mean RMSE and a 33-79% reduction in the upper tail of the 90% Highest Density Interval (HDI) over 500 Monte Carl (MC) trials across all time steps. Code will be made publicly available upon acceptance., Comment: Submitted to IEEE Aerospace and Electronic Systems

This study investigates whether Compressed-Language Models (CLMs), i.e. language models operating on raw byte streams from Compressed File Formats~(CFFs), can understand files compressed by CFFs. We focus on the JPEG format as a representative CFF, given its commonality and its representativeness of key concepts in compression, such as entropy coding and run-length encoding. We test if CLMs understand the JPEG format by probing their capabilities to perform along three axes: recognition of inherent file properties, handling of files with anomalies, and generation of new files. Our findings demonstrate that CLMs can effectively perform these tasks. These results suggest that CLMs can understand the semantics of compressed data when directly operating on the byte streams of files produced by CFFs. The possibility to directly operate on raw compressed files offers the promise to leverage some of their remarkable characteristics, such as their ubiquity, compactness, multi-modality and segment-nature.

Foundational generative models should be traceable to protect their owners and facilitate safety regulation. To achieve this, traditional approaches embed identifiers based on supervisory trigger-response signals, which are commonly known as backdoor watermarks. They are prone to failure when the model is fine-tuned with nontrigger data. Our experiments show that this vulnerability is due to energetic changes in only a few 'busy' layers during fine-tuning. This yields a novel arbitrary-in-arbitrary-out (AIAO) strategy that makes watermarks resilient to fine-tuning-based removal. The trigger-response pairs of AIAO samples across various neural network depths can be used to construct watermarked subpaths, employing Monte Carlo sampling to achieve stable verification results. In addition, unlike the existing methods of designing a backdoor for the input/output space of diffusion models, in our method, we propose to embed the backdoor into the feature space of sampled subpaths, where a mask-controlled trigger function is proposed to preserve the generation performance and ensure the invisibility of the embedded backdoor. Our empirical studies on the MS-COCO, AFHQ, LSUN, CUB-200, and DreamBooth datasets confirm the robustness of AIAO; while the verification rates of other trigger-based methods fall from ~90% to ~70% after fine-tuning, those of our method remain consistently above 90%.

In the era of the Internet of Things (IoT), objects connect through a dynamic network, empowered by technologies like 5G, enabling real-time data sharing. However, smart objects, notably autonomous vehicles, face challenges in critical local computations due to limited resources. Lightweight AI models offer a solution but struggle with diverse data distributions. To address this limitation, we propose a novel Multi-Stream Cellular Test-Time Adaptation (MSC-TTA) setup where models adapt on the fly to a dynamic environment divided into cells. Then, we propose a real-time adaptive student-teacher method that leverages the multiple streams available in each cell to quickly adapt to changing data distributions. We validate our methodology in the context of autonomous vehicles navigating across cells defined based on location and weather conditions. To facilitate future benchmarking, we release a new multi-stream large-scale synthetic semantic segmentation dataset, called DADE, and show that our multi-stream approach outperforms a single-stream baseline. We believe that our work will open research opportunities in the IoT and 5G eras, offering solutions for real-time model adaptation.

Understanding videos that contain multiple modalities is crucial, especially in egocentric videos, where combining various sensory inputs significantly improves tasks like action recognition and moment localization. However, real-world applications often face challenges with incomplete modalities due to privacy concerns, efficiency needs, or hardware issues. Current methods, while effective, often necessitate retraining the model entirely to handle missing modalities, making them computationally intensive, particularly with large training datasets. In this study, we propose a novel approach to address this issue at test time without requiring retraining. We frame the problem as a test-time adaptation task, where the model adjusts to the available unlabeled data at test time. Our method, MiDl~(Mutual information with self-Distillation), encourages the model to be insensitive to the specific modality source present during testing by minimizing the mutual information between the prediction and the available modality. Additionally, we incorporate self-distillation to maintain the model's original performance when both modalities are available. MiDl represents the first self-supervised, online solution for handling missing modalities exclusively at test time. Through experiments with various pretrained models and datasets, MiDl demonstrates substantial performance improvement without the need for retraining.

In this paper, we prove a global bifurcation result for the existence of non-radial branches of solutions to the paramterized family of $\Gamma$-symmetric problems $-\Delta u=f(\alpha,z,u)$, $u|_{\partial D}=0$ on the unit disc $D:=\{z\in \mathbb{C} : |z|<1\}$ with $u(z)\in \mathbb{R}^k$, where $\mathbb{R}^k$ is an orthogonal $\Gamma$-representation, $f: \mathbb{R} \times \overline{D} \times \mathbb{R}^k\to \mathbb{R}^k$ is a sub-linear $\Gamma$-equivariant continuous function, differentiable with respect to $u$ at zero and satisfying the conditions $f(\alpha, e^{i\theta}z,u)=f(\alpha, z,u)$ for all $\theta\in \mathbb{R}$ and $f(z,-u)=-f(z,u)$., Comment: updated version with fixed typos

We propose and study a realistic Continual Learning (CL) setting where learning algorithms are granted a restricted computational budget per time step while training. We apply this setting to large-scale semi-supervised Continual Learning scenarios with sparse label rates. Previous proficient CL methods perform very poorly in this challenging setting. Overfitting to the sparse labeled data and insufficient computational budget are the two main culprits for such a poor performance. Our new setting encourages learning methods to effectively and efficiently utilize the unlabeled data during training. To that end, we propose a simple but highly effective baseline, DietCL, which utilizes both unlabeled and labeled data jointly. DietCL meticulously allocates computational budget for both types of data. We validate our baseline, at scale, on several datasets, e.g., CLOC, ImageNet10K, and CGLM, under constraint budget setups. DietCL outperforms, by a large margin, all existing supervised CL algorithms as well as more recent continual semi-supervised methods. Our extensive analysis and ablations demonstrate that DietCL is stable under a full spectrum of label sparsity, computational budget, and various other ablations.

Tracking and identifying athletes on the pitch holds a central role in collecting essential insights from the game, such as estimating the total distance covered by players or understanding team tactics. This tracking and identification process is crucial for reconstructing the game state, defined by the athletes' positions and identities on a 2D top-view of the pitch, (i.e. a minimap). However, reconstructing the game state from videos captured by a single camera is challenging. It requires understanding the position of the athletes and the viewpoint of the camera to localize and identify players within the field. In this work, we formalize the task of Game State Reconstruction and introduce SoccerNet-GSR, a novel Game State Reconstruction dataset focusing on football videos. SoccerNet-GSR is composed of 200 video sequences of 30 seconds, annotated with 9.37 million line points for pitch localization and camera calibration, as well as over 2.36 million athlete positions on the pitch with their respective role, team, and jersey number. Furthermore, we introduce GS-HOTA, a novel metric to evaluate game state reconstruction methods. Finally, we propose and release an end-to-end baseline for game state reconstruction, bootstrapping the research on this task. Our experiments show that GSR is a challenging novel task, which opens the field for future research. Our dataset and codebase are publicly available at https://github.com/SoccerNet/sn-gamestate.

The escalating sophistication and volume of cyber threats in cloud environments necessitate a paradigm shift in strategies. Recognising the need for an automated and precise response to cyber threats, this research explores the application of AI and ML and proposes an AI-powered cyber incident response system for cloud environments. This system, encompassing Network Traffic Classification, Web Intrusion Detection, and post-incident Malware Analysis (built as a Flask application), achieves seamless integration across platforms like Google Cloud and Microsoft Azure. The findings from this research highlight the effectiveness of the Random Forest model, achieving an accuracy of 90% for the Network Traffic Classifier and 96% for the Malware Analysis Dual Model application. Our research highlights the strengths of AI-powered cyber security. The Random Forest model excels at classifying cyber threats, offering an efficient and robust solution. Deep learning models significantly improve accuracy, and their resource demands can be managed using cloud-based TPUs and GPUs. Cloud environments themselves provide a perfect platform for hosting these AI/ML systems, while container technology ensures both efficiency and scalability. These findings demonstrate the contribution of the AI-led system in guaranteeing a robust and scalable cyber incident response solution in the cloud.

The rapid advancement of artificial intelligence has led to significant improvements in automated decision-making. However, the increased performance of models often comes at the cost of explainability and transparency of their decision-making processes. In this paper, we investigate the capabilities of large language models to explain decisions, using football refereeing as a testing ground, given its decision complexity and subjectivity. We introduce the Explainable Video Assistant Referee System, X-VARS, a multi-modal large language model designed for understanding football videos from the point of view of a referee. X-VARS can perform a multitude of tasks, including video description, question answering, action recognition, and conducting meaningful conversations based on video content and in accordance with the Laws of the Game for football referees. We validate X-VARS on our novel dataset, SoccerNet-XFoul, which consists of more than 22k video-question-answer triplets annotated by over 70 experienced football referees. Our experiments and human study illustrate the impressive capabilities of X-VARS in interpreting complex football clips. Furthermore, we highlight the potential of X-VARS to reach human performance and support football referees in the future.

Recent advancements in diffusion models have shown remarkable proficiency in editing 2D images based on text prompts. However, extending these techniques to edit scenes in Neural Radiance Fields (NeRF) is complex, as editing individual 2D frames can result in inconsistencies across multiple views. Our crucial insight is that a NeRF scene's geometry can serve as a bridge to integrate these 2D edits. Utilizing this geometry, we employ a depth-conditioned ControlNet to enhance the coherence of each 2D image modification. Moreover, we introduce an inpainting approach that leverages the depth information of NeRF scenes to distribute 2D edits across different images, ensuring robustness against errors and resampling challenges. Our results reveal that this methodology achieves more consistent, lifelike, and detailed edits than existing leading methods for text-driven NeRF scene editing., Comment: 3D Scene Editing, Neural Rendering, Diffusion Models, Accepted to ECCV24

Movie trailers are an essential tool for promoting films and attracting audiences. However, the process of creating trailers can be time-consuming and expensive. To streamline this process, we propose an automatic trailer generation framework that generates plausible trailers from a full movie by automating shot selection and composition. Our approach draws inspiration from machine translation techniques and models the movies and trailers as sequences of shots, thus formulating the trailer generation problem as a sequence-to-sequence task. We introduce Trailer Generation Transformer (TGT), a deep-learning framework utilizing an encoder-decoder architecture. TGT movie encoder is tasked with contextualizing each movie shot representation via self-attention, while the autoregressive trailer decoder predicts the feature representation of the next trailer shot, accounting for the relevance of shots' temporal order in trailers. Our TGT significantly outperforms previous methods on a comprehensive suite of metrics., Comment: Accepted to CVPR 2024

In this paper, we investigate joint 3-dimensional (3D) trajectory planning and resource allocation for rotary-wing unmanned aerial vehicle (UAV) synthetic aperture radar (SAR) sensing. To support emerging real-time SAR applications and enable live mission control, we incorporate real-time communication with a ground station (GS). The UAV's main mission is the mapping of large areas of interest (AoIs) using an onboard SAR system and transferring the unprocessed raw radar data to the ground in real time. We propose a robust trajectory and resource allocation design that takes into account random UAV trajectory deviations. To this end, we model the UAV trajectory deviations and study their effect on the radar coverage. Then, we formulate a robust non-convex mixed-integer non-linear program (MINLP) such that the UAV 3D trajectory and resources are jointly optimized for maximization of the radar ground coverage. A low-complexity sub-optimal solution for the formulated problem is presented. Furthermore, to assess the performance of the sub-optimal algorithm, we derive an upper bound on the optimal solution based on monotonic optimization theory. Simulation results show that the proposed sub-optimal algorithm achieves close-to-optimal performance and not only outperforms several benchmark schemes but is also robust with respect to UAV trajectory deviations.

The modern surge in camera usage alongside widespread computer vision technology applications poses significant privacy and security concerns. Current artificial intelligence (AI) technologies aid in recognizing relevant events and assisting in daily tasks in homes, offices, hospitals, etc. The need to access or process personal information for these purposes raises privacy concerns. While software-level solutions like face de-identification provide a good privacy/utility trade-off, they present vulnerabilities to sniffing attacks. In this paper, we propose a hardware-level face de-identification method to solve this vulnerability. Specifically, our approach first learns an optical encoder along with a regression model to obtain a face heatmap while hiding the face identity from the source image. We also propose an anonymization framework that generates a new face using the privacy-preserving image, face heatmap, and a reference face image from a public dataset as input. We validate our approach with extensive simulations and hardware experiments., Comment: Accepted to CVPR 2024. Project Website and Code coming soon

Self-supervised pre-training of image encoders is omnipresent in the literature, particularly following the introduction of Masked autoencoders (MAE). Current efforts attempt to learn object-centric representations from motion in videos. In particular, SiamMAE recently introduced a Siamese network, training a shared-weight encoder from two frames of a video with a high asymmetric masking ratio (95%). In this work, we propose CropMAE, an alternative approach to the Siamese pre-training introduced by SiamMAE. Our method specifically differs by exclusively considering pairs of cropped images sourced from the same image but cropped differently, deviating from the conventional pairs of frames extracted from a video. CropMAE therefore alleviates the need for video datasets, while maintaining competitive performances and drastically reducing pre-training and learning time. Furthermore, we demonstrate that CropMAE learns similar object-centric representations without explicit motion, showing that current self-supervised learning methods do not learn such representations from explicit object motion, but rather thanks to the implicit image transformations that occur between the two views. Finally, CropMAE achieves the highest masking ratio to date (98.5%), enabling the reconstruction of images using only two visible patches. Our code is available at https://github.com/alexandre-eymael/CropMAE., Comment: 19 pages, 7 figures, 5 tables, 3 pages of supplementary material. Paper accepted at ECCV 2024