Two summers ago, the Santa Monica-based company behind the popular video game "Call of Duty" sent a letter to a 24-year-old man in Antioch, Tenn., who went by the online handle "Lerggy." Known in real life as Ryan Rothholz, court filings say, he is the creator of "Lergware," hacking software that enabled Call of Duty players to cheat by kicking opponents offline. A lawsuit filed in May against Rothholz and others allegedly involved in the hacking scheme is the latest salvo in years-long campaign by Activision-Blizzard and other companies to rid their games of cheating. The war is being waged in the Central District of California civil courts, but the defendants are scattered across the country and as far away as Australia. An immersive "first-person shooter" game, Call of Duty takes players into simulated, realistic military combat.

Learning from non-stationary data streams subject to concept drift requires models that can adapt on-the-fly while remaining resource-efficient. Existing adaptive ensemble methods often rely on coarse-grained adaptation mechanisms or simple voting schemes that fail to optimally leverage specialized knowledge. This paper introduces DriftMoE, an online Mixture-of-Experts (MoE) architecture that addresses these limitations through a novel co-training framework. DriftMoE features a compact neural router that is co-trained alongside a pool of incremental Hoeffding tree experts. The key innovation lies in a symbiotic learning loop that enables expert specialization: the router selects the most suitable expert for prediction, the relevant experts update incrementally with the true label, and the router refines its parameters using a multi-hot correctness mask that reinforces every accurate expert. This feedback loop provides the router with a clear training signal while accelerating expert specialization. We evaluate DriftMoE's performance across nine state-of-the-art data stream learning benchmarks spanning abrupt, gradual, and real-world drifts testing two distinct configurations: one where experts specialize on data regimes (multi-class variant), and another where they focus on single-class specialization (task-based variant). Our results demonstrate that DriftMoE achieves competitive results with state-of-the-art stream learning adaptive ensembles, offering a principled and efficient approach to concept drift adaptation.

Eye-tracking data reveals valuable insights into users' cognitive states but is difficult to analyze due to its structured, non-linguistic nature. While large language models (LLMs) excel at reasoning over text, they struggle with temporal and numerical data. This paper presents a multimodal human-AI collaborative framework designed to enhance cognitive pattern extraction from eye-tracking signals. The framework includes: (1) a multi-stage pipeline using horizontal and vertical segmentation alongside LLM reasoning to uncover latent gaze patterns; (2) an Expert-Model Co-Scoring Module that integrates expert judgment with LLM output to generate trust scores for behavioral interpretations; and (3) a hybrid anomaly detection module combining LSTM-based temporal modeling with LLM-driven semantic analysis. Our results across several LLMs and prompt strategies show improvements in consistency, interpretability, and performance, with up to 50% accuracy in difficulty prediction tasks. This approach offers a scalable, interpretable solution for cognitive modeling and has broad potential in adaptive learning, human-computer interaction, and educational analytics.

The rapid development of COVID-19 vaccines has showcased the global community's ability to combat infectious diseases. However, the need for post-licensure surveillance systems has grown due to the limited window for safety data collection in clinical trials and early widespread implementation. This study aims to employ Natural Language Processing (NLP) techniques and Active Learning (AL) to rapidly develop a classifier that detects potential vaccine safety issues from emergency department (ED) notes. ED triage notes, containing expert, succinct vital patient information at the point of entry to health systems, can significantly contribute to timely vaccine safety signal surveillance. While keyword-based classification can be effective, it may yield false positives and demand extensive keyword modifications. This is exacerbated by the infrequency of vaccination-related ED presentations and their similarity to other reasons for ED visits. NLP offers a more accurate and efficient alternative, albeit requiring annotated data, which is often scarce in the medical field. Active learning optimizes the annotation process and the quality of annotated data, which can result in faster model implementation and improved model performance. This work combines active learning, data augmentation, and active learning and evaluation techniques to create a classifier that is used to enhance vaccine safety surveillance from ED triage notes.

In this paper we propose a modular nonlinear least squares filtering approach for systems composed of independent subsystems. The state and error covariance estimate of each subsystem is updated independently, even when a relative measurement simultaneously depends on the states of multiple subsystems. We integrate the Covariance Intersection (CI) algorithm as part of our solution in order to prevent double counting of information when subsystems share estimates with each other. An alternative derivation of the CI algorithm based on least squares estimation makes this integration possible. We particularise the proposed approach to the robot-landmark localization problem. In this problem, noisy measurements of the bearing angle to a stationary landmark position measured relative to the SE(2) pose of a moving robot couple the estimation problems for the robot pose and the landmark position. In a randomized simulation study, we benchmark the proposed modular method against a monolithic joint state filter to elucidate their respective trade-offs. In this study we also include variants of the proposed method that achieve a graceful degradation of performance with reduced communication and bandwidth requirements.

The massive increase in data in scientific research requires the development and application of robust tools for data analysis and m achine l earning (ML) that are findable, accessible, interoperable, re usable (FAIR) and interpretable. In domains, such as b iomaterials s cience, e ngineering, c hemistry, h ealthcare and b io sciences, data - driven discovery typically requires interdisciplinary teams . These teams collaborate to implement unbiased data pre - processing strategies, select appropriate modelling techniques, and interpret model outputs to accelerate and inform research outcomes and support rational design and decision - making. This process is often iterative, with experts providing feedback over long periods of time to refine models and optimise the methodology adopted . In cases where initial analysis identifies issues with the data, such as outliers, unbalance d data classes, or experimental measurement uncertainty, another round of data collection and pre - processing might be necessary . That means that data for the same problem are likely to be analysed multiple times using different dataset versions and methodological pipelines. For interdisciplinary co - development of analytic s, there is also a need for tools that allow domain experts to focus on interpreting and using analysis results, rather than developing code . The widespread use of ML and the overwhelming availability of thousands of community - driven open - source packages in Python and R increases the barrier for interoperable and reusable data analysis methodologies . To facilitate accurate analy tics, transparency, and modelling results comparison, there is a strong need for easy - to - use tools that automatically track data, all methodological choices, performance metrics, and corresponding results.

Robustness and resource-efficiency are two highly desirable properties for modern machine learning models. However, achieving them jointly remains a challenge. In this paper, we position high learning rates as a facilitator for simultaneously achieving robustness to spurious correlations and network compressibility. We demonstrate that large learning rates also produce desirable representation properties such as invariant feature utilization, class separation, and activation sparsity. Importantly, our findings indicate that large learning rates compare favorably to other hyperparameters and regularization methods, in consistently satisfying these properties in tandem. In addition to demonstrating the positive effect of large learning rates across diverse spurious correlation datasets, models, and optimizers, we also present strong evidence that the previously documented success of large learning rates in standard classification tasks is likely due to its effect on addressing hidden/rare spurious correlations in the training dataset.

In the design of customer relationship management (CRM) systems, accurately identifying customer types and offering personalized services are key to enhancing customer satisfaction and loyalty. However, this process faces the challenge of discerning customer voices and intentions, and general pre-trained automatic speech recognition (ASR) models make it difficult to effectively address industry-specific speech recognition tasks. To address this issue, we innovatively proposed a solution for fine-tuning industry-specific ASR models, which significantly improved the performance of the fine-tuned ASR models in industry applications. Experimental results show that our method substantially improves the crucial auxiliary role of the ASR model in industry CRM systems, and this approach has also been adopted in actual industrial applications.

Geo-localization from a single image at planet scale (essentially an advanced or extreme version of the kidnapped robot problem) is a fundamental and challenging task in applications such as navigation, autonomous driving and disaster response due to the vast diversity of locations, environmental conditions, and scene variations. Traditional retrieval-based methods for geo-localization struggle with scalability and perceptual aliasing, while classification-based approaches lack generalization and require extensive training data. Recent advances in vision-language models (VLMs) offer a promising alternative by leveraging contextual understanding and reasoning. However, while VLMs achieve high accuracy, they are often prone to hallucinations and lack interpretability, making them unreliable as standalone solutions. In this work, we propose a novel hybrid geo-localization framework that combines the strengths of VLMs with retrieval-based visual place recognition (VPR) methods. Our approach first leverages a VLM to generate a prior, effectively guiding and constraining the retrieval search space. We then employ a retrieval step, followed by a re-ranking mechanism that selects the most geographically plausible matches based on feature similarity and proximity to the initially estimated coordinates. We evaluate our approach on multiple geo-localization benchmarks and show that it consistently outperforms prior state-of-the-art methods, particularly at street (up to 4.51%) and city level (up to 13.52%). Our results demonstrate that VLM-generated geographic priors in combination with VPR lead to scalable, robust, and accurate geo-localization systems.

The vulnerability of neural networks to adversarial perturbations has necessitated formal verification techniques that can rigorously certify the quality of neural networks. As the state-of-the-art, branch and bound (BaB) is a "divide-and-conquer" strategy that applies off-the-shelf verifiers to sub-problems for which they perform better. While BaB can identify the sub-problems that are necessary to be split, it explores the space of these sub-problems in a naive "first-come-first-serve" manner, thereby suffering from an issue of inefficiency to reach a verification conclusion. To bridge this gap, we introduce an order over different sub-problems produced by BaB, concerning with their different likelihoods of containing counterexamples. Based on this order, we propose a novel verification framework Oliva that explores the sub-problem space by prioritizing those sub-problems that are more likely to find counterexamples, in order to efficiently reach the conclusion of the verification. Even if no counterexample can be found in any sub-problem, it only changes the order of visiting different sub-problem and so will not lead to a performance degradation. Specifically, Oliva has two variants, including $Oliva^{GR}$, a greedy strategy that always prioritizes the sub-problems that are more likely to find counterexamples, and $Oliva^{SA}$, a balanced strategy inspired by simulated annealing that gradually shifts from exploration to exploitation to locate the globally optimal sub-problems. We experimentally evaluate the performance of Oliva on 690 verification problems spanning over 5 models with datasets MNIST and CIFAR10. Compared to the state-of-the-art approaches, we demonstrate the speedup of Oliva for up to 25X in MNIST, and up to 80X in CIFAR10.