Quantum Software Engineering (QSE) is a research area practiced by tech firms. Quantum developers face challenges in optimizing quantum computing and QSE concepts. They use Stack Overflow (SO) to discuss challenges and label posts with specialized quantum tags, which often refer to technical aspects rather than developer posts. Categorizing questions based on quantum concepts can help identify frequent QSE challenges. We conducted studies to classify questions into various challenges. We extracted 2829 questions from Q&A platforms using quantum-related tags. Posts were analyzed to identify frequent challenges and develop a novel grounded theory. Challenges include Tooling, Theoretical, Learning, Conceptual, Errors, and API Usage. Through content analysis and grounded theory, discussions were annotated with common challenges to develop a ground truth dataset. ChatGPT validated human annotations and resolved disagreements. Fine-tuned transformer algorithms, including BERT, DistilBERT, and RoBERTa, classified discussions into common challenges. We achieved an average accuracy of 95% with BERT DistilBERT, compared to fine-tuned Deep and Machine Learning (D&ML) classifiers, including Feedforward Neural Networks (FNN), Convolutional Neural Networks (CNN), and Long Short-Term Memory networks (LSTM), which achieved accuracies of 89%, 86%, and 84%, respectively. The Transformer-based approach outperforms the D&ML-based approach with a 6\% increase in accuracy by processing actual discussions, i.e., without data augmentation. We applied SHAP (SHapley Additive exPlanations) for model interpretability, revealing how linguistic features drive predictions and enhancing transparency in classification. These findings can help quantum vendors and forums better organize discussions for improved access and readability. However,empirical evaluation studies with actual developers and vendors are needed.

Deep neural network predictions are notoriously difficult to interpret. Feature attribution methods aim to explain these predictions by identifying the contribution of each input feature. Faithfulness, often evaluated using the area over the perturbation curve (AOPC), reflects feature attributions' accuracy in describing the internal mechanisms of deep neural networks. However, many studies rely on AOPC to compare faithfulness across different models, which we show can lead to false conclusions about models' faithfulness. Specifically, we find that AOPC is sensitive to variations in the model, resulting in unreliable cross-model comparisons. Moreover, AOPC scores are difficult to interpret in isolation without knowing the model-specific lower and upper limits. To address these issues, we propose a normalization approach, Normalized AOPC (NAOPC), enabling consistent cross-model evaluations and more meaningful interpretation of individual scores. Our experiments demonstrate that this normalization can radically change AOPC results, questioning the conclusions of earlier studies and offering a more robust framework for assessing feature attribution faithfulness. Our code is available at this https URL

Ensuring faithful interpretability in large language models is imperative for trustworthy and reliable AI. A key obstacle is self-repair, a phenomenon where networks compensate for reduced signal in one component by amplifying others, masking the true importance of the ablated component. While prior work attributes self-repair to layer normalization and back-up components that compensate for ablated components, we identify a novel form occurring within the attention mechanism, where softmax redistribution conceals the influence of important attention scores. This leads traditional ablation and gradient-based methods to underestimate the significance of all components contributing to these attention scores. We introduce Gradient Interaction Modifications (GIM), a technique that accounts for self-repair during backpropagation. Extensive experiments across multiple large language models (Gemma 2B/9B, LLAMA 1B/3B/8B, Qwen 1.5B/3B) and diverse tasks demonstrate that GIM significantly improves faithfulness over existing circuit identification and feature attribution methods. Our work is a significant step toward better understanding the inner mechanisms of LLMs, which is crucial for improving them and ensuring their safety. Our code is available at this https URL.

Reconstructing an image from its Radon transform is a fundamental computed tomography (CT) task arising in applications such as X-ray scans. In many practical scenarios, a full 180-degree scan is not feasible, or there is a desire to reduce radiation exposure. In these limited-angle settings, the problem becomes ill-posed, and methods designed for full-view data often leave significant artifacts. We propose a very low-data approach to reconstruct the original image from its Radon transform under severe angle limitations. Because the inverse problem is ill-posed, we combine multiple regularization methods, including Total Variation, a sinogram filter, Deep Image Prior, and a patch-level autoencoder. We use a differentiable implementation of the Radon transform, which allows us to use gradient-based techniques to solve the inverse problem. Our method is evaluated on a dataset from the Helsinki Tomography Challenge 2022, where the goal is to reconstruct a binary disk from its limited-angle sinogram. We only use a total of 12 data points--eight for learning a prior and four for hyperparameter selection--and achieve results comparable to the best synthetic data-driven approaches.

Context: Tangled commits are changes to software that address multiple concerns at once. For researchers interested in bugs, tangled commits mean that they actually study not only bugs, but also other concerns irrelevant for the study of bugs. Objective: We want to improve our understanding of the prevalence of tangling and the types of changes that are tangled within bug fixing commits. Methods: We use a crowd sourcing approach for manual labeling to validate which changes contribute to bug fixes for each line in bug fixing commits. Each line is labeled by four participants. If at least three participants agree on the same label, we have consensus. Results: We estimate that between 17% and 32% of all changes in bug fixing commits modify the source code to fix the underlying problem. However, when we only consider changes to the production code files this ratio increases to 66% to 87%. We find that about 11% of lines are hard to label leading to active disagreements between participants. Due to confirmed tangling and the uncertainty in our data, we estimate that 3% to 47% of data is noisy without manual untangling, depending on the use case. Conclusion: Tangled commits have a high prevalence in bug fixes and can lead to a large amount of noise in the data. Prior research indicates that this noise may alter results. As researchers, we should be skeptics and assume that unvalidated data is likely very noisy, until proven otherwise.

Generative AI offers vast opportunities for creating visualisations, such as graphics, videos, and images. However, recent studies around AI-generated visualisations have primarily focused on the creation process and image quality, overlooking representational biases. This study addresses this gap by testing representation biases in AI-generated pictures in an occupational setting and evaluating how two AI image generator tools, DALL-E 3 and Ideogram, compare. Additionally, the study discusses topics such as ageing and emotions in AI-generated images. As AI image tools are becoming more widely used, addressing and mitigating harmful gender biases becomes essential to ensure diverse representation in media and professional settings. In this study, over 750 AI-generated images of occupations were prompted. The thematic analysis results revealed that both DALL-E 3 and Ideogram reinforce traditional gender stereotypes in AI-generated images, although to varying degrees. These findings emphasise that AI visualisation tools risk reinforcing narrow representations. In our discussion section, we propose suggestions for practitioners, individuals and researchers to increase representation when generating images with visible genders.

Unsupervised learning of disentangled representations has been closely tied to enhancing the representation intepretability of Recommender Systems (RSs). This has been achieved by making the representation of individual features more distinctly separated, so that it is easier to attribute the contribution of features to the model's predictions. However, such advantages in interpretability and feature attribution have mainly been explored qualitatively. Moreover, the effect of disentanglement on the model's recommendation performance has been largely overlooked. In this work, we reproduce the recommendation performance, representation disentanglement and representation interpretability of five well-known recommendation models on four RS datasets. We quantify disentanglement and investigate the link of disentanglement with recommendation effectiveness and representation interpretability. While several existing work in RSs have proposed disentangled representations as a gateway to improved effectiveness and interpretability, our findings show that disentanglement is not necessarily related to effectiveness but is closely related to representation interpretability. Our code and results are publicly available at this https URL.

We consider the problem of recovering a mental target (e.g., an image of a face) that a participant has in mind from paired EEG (i.e., brain responses) and image (i.e., perceived faces) data collected during interactive sessions without access to labeled information. The problem has been previously explored with labeled data but not via self-calibration, where labeled data is unavailable. Here, we present the first framework and an algorithm, CURSOR, that learns to recover unknown mental targets without access to labeled data or pre-trained decoders. Our experiments on naturalistic images of faces demonstrate that CURSOR can (1) predict image similarity scores that correlate with human perceptual judgments without any label information, (2) use these scores to rank stimuli against an unknown mental target, and (3) generate new stimuli indistinguishable from the unknown mental target (validated via a user study, N=53).

Drones have recently emerged as a faster, safer, and cost-efficient way for last-mile deliveries of parcels, particularly for urgent medical deliveries highlighted during the pandemic. This paper addresses a new challenge of multi-parcel delivery with a swarm of energy-aware drones, accounting for time-sensitive customer requirements. Each drone plans an optimal multi-parcel route within its battery-restricted flight range to minimize delivery delays and reduce energy consumption. The problem is tackled by decomposing it into three sub-problems: (1) optimizing depot locations and service areas using K-means clustering; (2) determining the optimal flight range for drones through reinforcement learning; and (3) planning and selecting multi-parcel delivery routes via a new optimized plan selection approach. To integrate these solutions and enhance long-term efficiency, we propose a novel algorithm leveraging actor-critic-based multi-agent deep reinforcement learning. Extensive experimentation using realistic delivery datasets demonstrate an exceptional performance of the proposed algorithm. We provide new insights into economic efficiency (minimize energy consumption), rapid operations (reduce delivery delays and overall execution time), and strategic guidance on depot deployment for practical logistics applications.

Multivariate Statistical Process Control (MSPC) is a framework for monitoring and diagnosing complex processes by analysing the relationships between multiple process variables simultaneously. Kernel MSPC extends the methodology by leveraging kernel functions to capture non-linear relationships between the data, enhancing the process monitoring capabilities. However, optimising the kernel MSPC parameters, such as the kernel type and kernel parameters, is often done in literature in time-consuming and non-procedural manners such as cross-validation or grid search. In the present paper, we propose optimising the kernel MSPC parameters with Kernel Flows (KF), a recent kernel learning methodology introduced for Gaussian Process Regression (GPR). Apart from the optimisation technique, the novelty of the study resides also in the utilisation of kernel combinations for learning the optimal kernel type, and introduces individual kernel parameters for each variable. The proposed methodology is evaluated with multiple cases from the benchmark Tennessee Eastman Process. The faults are detected for all evaluated cases, including the ones not detected in the original study.

In the domain of rotating machinery, bearings are vulnerable to different mechanical faults, including ball, inner, and outer race faults. Various techniques can be used in condition-based monitoring, from classical signal analysis to deep learning methods. Based on the complex working conditions of rotary machines, multivariate statistical process control charts such as Hotelling's $T^2$ and Squared Prediction Error are useful for providing early warnings. However, these methods are rarely applied to condition monitoring of rotating machinery due to the univariate nature of the datasets. In the present paper, we propose a multivariate statistical process control-based fault detection method that utilizes multivariate data composed of Fourier transform features extracted for fixed-time batches. Our approach makes use of the multidimensional nature of Fourier transform characteristics, which record more detailed information about the machine's status, in an effort to enhance early defect detection and diagnosis. Experiments with varying vibration measurement locations (Fan End, Drive End), fault types (ball, inner, and outer race faults), and motor loads (0-3 horsepower) are used to validate the suggested approach. The outcomes illustrate our method's effectiveness in fault detection and point to possible broader uses in industrial maintenance.

With unorganized, unplanned and improper use of limited raw materials, an abundant amount of waste is being produced, which is harmful to our environment and ecosystem. While traditional linear production lines fail to address far-reaching issues like waste production and a shorter product life cycle, a prospective concept, namely circular economy (CE), has shown promising prospects to be adopted at industrial and governmental levels. CE aims to complete the product life cycle loop by bringing out the highest values from raw materials in the design phase and later on by reusing, recycling, and remanufacturing. Innovative technologies like artificial intelligence (AI) and machine learning(ML) provide vital assistance in effectively adopting and implementing CE in real-world practices. This study explores the adoption and integration of applied AI techniques in CE. First, we conducted bibliometric analysis on a collection of 104 SCOPUS indexed documents exploring the critical research criteria in AI and CE. Forty papers were picked to conduct a systematic literature review from these documents. The selected documents were further divided into six categories: sustainable development, reverse logistics, waste management, supply chain management, recycle & reuse, and manufacturing development. Comprehensive research insights and trends have been extracted and delineated. Finally, the research gap needing further attention has been identified and the future research directions have also been discussed.

Language Model (LM) pruning compresses the model by removing weights, nodes, or other parts of its architecture. Typically, pruning focuses on the resulting efficiency gains at the cost of effectiveness. However, when looking at how individual data points are affected by pruning, it turns out that a particular subset of data points always bears most of the brunt (in terms of reduced accuracy) when pruning, but this effect goes unnoticed when reporting the mean accuracy of all data points. These data points are called PIEs and have been studied in image processing, but not in NLP. In a study of various NLP datasets, pruning methods, and levels of compression, we find that PIEs impact inference quality considerably, regardless of class frequency, and that BERT is more prone to this than BiLSTM. We also find that PIEs contain a high amount of data points that have the largest influence on how well the model generalises to unseen data. This means that when pruning, with seemingly moderate loss to accuracy across all data points, we in fact hurt tremendously those data points that matter the most. We trace what makes PIEs both hard and impactful to inference to their overall longer and more semantically complex text. These findings are novel and contribute to understanding how LMs are affected by pruning. The code is available at: this https URL

Federated learning (FL) is a popular approach that enables organizations to train machine learning models without compromising data privacy and security. As the field of FL continues to grow, it is crucial to have a thorough understanding of the topic, current trends and future research directions for information systems (IS) researchers. Consequently, this paper conducts a comprehensive computational literature review on FL and presents the research landscape. By utilizing advanced data analytics and leveraging the topic modeling approach, we identified and analyzed the most prominent 15 topics and areas that have influenced the research on FL. We also proposed guiding research questions to stimulate further research directions for IS scholars. Our work is valuable for scholars, practitioners, and policymakers since it offers a comprehensive overview of state-of-the-art research on FL.

Bayesian optimal experimental design (OED) provides a principled framework for selecting the most informative observational settings in experiments. With rapid advances in computational power, Bayesian OED has become increasingly feasible for inference problems involving large-scale simulations, attracting growing interest in fields such as inverse problems. In this paper, we introduce a novel design criterion based on the expected Wasserstein-$p$ distance between the prior and posterior distributions. Especially, for $p=2$, this criterion shares key parallels with the widely used expected information gain (EIG), which relies on the Kullback--Leibler divergence instead. First, the Wasserstein-2 criterion admits a closed-form solution for Gaussian regression, a property which can be also leveraged for approximative schemes. Second, it can be interpreted as maximizing the information gain measured by the transport cost incurred when updating the prior to the posterior. Our main contribution is a stability analysis of the Wasserstein-1 criterion, where we provide a rigorous error analysis under perturbations of the prior or likelihood. We partially extend this study also to the Wasserstein-2 criterion. In particular, these results yield error rates when empirical approximations of priors are used. Finally, we demonstrate the computability of the Wasserstein-2 criterion and demonstrate our approximation rates through simulations.

In this paper, a hyperparameter tuning based Bayesian optimization of digital twins is carried out to diagnose various faults in grid connected inverters. As fault detection and diagnosis require very high precision, we channelize our efforts towards an online optimization of the digital twins, which, in turn, allows a flexible implementation with limited amount of data. As a result, the proposed framework not only becomes a practical solution for model versioning and deployment of digital twins design with limited data, but also allows integration of deep learning tools to improve the hyperparameter tuning capabilities. For classification performance assessment, we consider different fault cases in virtual synchronous generator (VSG) controlled grid-forming converters and demonstrate the efficacy of our approach. Our research outcomes reveal the increased accuracy and fidelity levels achieved by our digital twin design, overcoming the shortcomings of traditional hyperparameter tuning methods.

The work analyzes the effect of a magnetic field $B$ directed along the $c$ axis ($B \parallel c$) up to 9~T on the resistivity $\rho(T)$, fluctuation conductivity (FLC) $\sigma'(T)$ and pseudogap $\Delta^*(T)$ in thin films of YBa$_2$Cu$_3$O$_{7-\delta}$ with a critical temperature of the superconducting transition $T_c = 88.8$~K. In contrast to previous work, where the magnetic field was directed along the $ab$ plane ($B \parallel ab$), the influence of the field on the sample is stronger due to the contribution of both spin--orbit and Zeeman effects. It was found that the BEC--BCS transition temperature, $T_{\mathrm{pair}}$, which corresponds to the maximum of the $\Delta^*(T)$ dependence, shifts to the region of lower temperatures with increasing $B$, and the maximum value of $\Delta^*(T_{\mathrm{pair}})$ decreases in fields $B > 5$~T. It was found that with increasing field, the low-temperature maximum near $T_0$ is smeared and disappears at $B > 1$~T. In addition, above the Ginzburg temperature $T_G$, for $B > 1$~T, a minimum appears on $\Delta^*(T)$ at $T_{\min}$, which becomes very pronounced with a subsequent increase in $B$. As a result, the overall value of $\Delta^*(T_G)$ decreases noticeably, most likely due to the pair-breaking effect. At the same time, $\Delta T_{\mathrm{fl}}$ and $\xi_c(0)$ increase sharply by approximately 3 times with increasing $B$ above 1~T. Our results confirm the possibility of the formation of a vortex state in YBa$_2$Cu$_3$O$_{7-\delta}$ by a magnetic field and its evolution with increasing $B$.