Research Papers

Artificial intelligence (AI) has brought benefits, but it may also cause harm if it is not appropriately developed. Current development is mainly driven by a "technology-centered" approach, causing many failures. For example, the AI Incident Database has documented over a thousand AI-related accidents. To address these challenges, a human-centered AI (HCAI) approach has been promoted and has received a growing level of acceptance over the last few years. HCAI calls for combining AI with user experience (UX) design will enable the development of AI systems (e.g., autonomous vehicles, intelligent user interfaces, or intelligent decision-making systems) to achieve its design goals such as usable/explainable AI, human-controlled AI, and ethical AI. While HCAI promotion continues, it has not specifically addressed the collaboration between AI and human-computer interaction (HCI) communities, resulting in uncertainty about what action should be taken by both sides to apply HCAI in developing AI systems. This Viewpoint focuses on the collaboration between the AI and HCI communities, which leads to nine recommendations for effective collaboration to enable HCAI in developing AI systems.

Traffic accident anticipation is a vital function of Automated Driving Systems (ADSs) for providing a safety-guaranteed driving experience. An accident anticipation model aims to predict accidents promptly and accurately before they occur. Existing Artificial Intelligence (AI) models of accident anticipation lack a human-interpretable explanation of their decision-making. Although these models perform well, they remain a black-box to the ADS users, thus difficult to get their trust. To this end, this paper presents a Gated Recurrent Unit (GRU) network that learns spatio-temporal relational features for the early anticipation of traffic accidents from dashcam video data. A post-hoc attention mechanism named Grad-CAM is integrated into the network to generate saliency maps as the visual explanation of the accident anticipation decision. An eye tracker captures human eye fixation points for generating human attention maps. The explainability of network-generated saliency maps is evaluated in comparison to human attention maps. Qualitative and quantitative results on a public crash dataset confirm that the proposed explainable network can anticipate an accident on average 4.57 seconds before it occurs, with 94.02% average precision. In further, various post-hoc attention-based XAI methods are evaluated and compared. It confirms that the Grad-CAM chosen by this study can generate high-quality, human-interpretable saliency maps (with 1.23 Normalized Scanpath Saliency) for explaining the crash anticipation decision. Importantly, results confirm that the proposed AI model, with a human-inspired design, can outperform humans in the accident anticipation.

Strategies based on Explainable Artificial Intelligence - XAI have emerged in computing to promote a better understanding of predictions made by black box models. Most XAI measures used today explain these types of models, generating attribute rankings aimed at explaining the model, that is, the analysis of Attribute Importance of Model. There is no consensus on which XAI measure generates an overall explainability rank. For this reason, several proposals for tools have emerged (Ciu, Dalex, Eli5, Lofo, Shap and Skater). An experimental benchmark of explainable AI techniques capable of producing global explainability ranks based on tabular data related to different problems and ensemble models are presented herein. Seeking to answer questions such as "Are the explanations generated by the different measures the same, similar or different?" and "How does data complexity play along model explainability?" The results from the construction of 82 computational models and 592 ranks shed some light on the other side of the problem of explainability: dataset complexity!

In situations where explanations of black-box models may be useful, the fairness of the black-box is also often a relevant concern. However, the link between the fairness of the black-box model and the behavior of explanations for the black-box is unclear. We focus on explanations applied to tabular datasets, suggesting that explanations do not necessarily preserve the fairness properties of the black-box algorithm. In other words, explanation algorithms can ignore or obscure critical relevant properties, creating incorrect or misleading explanations. More broadly, we propose future research directions for evaluating and generating explanations such that they are informative and relevant from a fairness perspective.

Deep learning has now become the de facto approach to the recognition of anomalies in medical imaging. Their 'black box' way of classifying medical images into anomaly labels poses problems for their acceptance, particularly with clinicians. Current explainable AI methods offer justifications through visualizations such as heat maps but cannot guarantee that the network is focusing on the relevant image region fully containing the anomaly. In this paper, we develop an approach to explainable AI in which the anomaly is assured to be overlapping the expected location when present. This is made possible by automatically extracting location-specific labels from textual reports and learning the association of expected locations to labels using a hybrid combination of Bi-Directional Long Short-Term Memory Recurrent Neural Networks (Bi-LSTM) and DenseNet-121. Use of this expected location to bias the subsequent attention-guided inference network based on ResNet101 results in the isolation of the anomaly at the expected location when present. The method is evaluated on a large chest X-ray dataset.

The societal and ethical relevance of computational creativity

Michele Loi, Eleonora Viganò, Lonneke van der Plas

In this paper, we provide a philosophical account of the value of creative systems for individuals and society. We characterize creativity in very broad philosophical terms, encompassing natural, existential, and social creative processes, such as natural evolution and entrepreneurship, and explain why creativity understood in this way is instrumental for advancing human well-being in the long term. We then explain why current mainstream AI tends to be anti-creative, which means that there are moral costs of employing this type of AI in human endeavors, although computational systems that involve creativity are on the rise. In conclusion, there is an argument for ethics to be more hospitable to creativity-enabling AI, which can also be in a trade-off with other values promoted in AI ethics, such as its explainability and accuracy.

People often rely on online reviews to make purchase decisions. The present work aimed to gain an understanding of a machine learning model's prediction mechanism by visualizing the effect of sentiments extracted from online hotel reviews with explainable AI (XAI) methodology. Study 1 used the extracted sentiments as features to predict the review ratings by five machine learning algorithms (knn, CART decision trees, support vector machines, random forests, gradient boosting machines) and identified random forests as best algorithm. Study 2 analyzed the random forests model by feature importance and revealed the sentiments joy, disgust, positive and negative as the most predictive features. Furthermore, the visualization of additive variable attributions and their prediction distribution showed correct prediction in direction and effect size for the 5-star rating but partially wrong direction and insufficient effect size for the 1-star rating. These prediction details were corroborated by a what-if analysis for the four top features. In conclusion, the prediction mechanism of a machine learning model can be uncovered by visualization of particular observations. Comparing instances of contrasting ground truth values can draw a differential picture of the prediction mechanism and inform decisions for model improvement.

Traditionally, researchers in automatic face recognition and biometric technologies have focused on developing accurate algorithms. With this technology being integrated into operational systems, engineers and scientists are being asked, do these systems meet societal norms? The origin of this line of inquiry is `trust' of artificial intelligence (AI) systems. In this paper, we concentrate on adapting explainable AI to face recognition and biometrics, and we present four principles of explainable AI to face recognition and biometrics. The principles are illustrated by $\it{four}$ case studies, which show the challenges and issues in developing algorithms that can produce explanations.

While Chain-of-Thought (CoT) monitoring offers a unique opportunity for AI safety, this opportunity could be lost through shifts in training practices or model architecture. To help preserve monitorability, we propose a pragmatic way to measure two components of it: legibility (whether the reasoning can be followed by a human) and coverage (whether the CoT contains all the reasoning needed for a human to also produce the final output). We implement these metrics with an autorater prompt that enables any capable LLM to compute the legibility and coverage of existing CoTs. After sanity-checking our prompted autorater with synthetic CoT degradations, we apply it to several frontier models on challenging benchmarks, finding that they exhibit high monitorability. We present these metrics, including our complete autorater prompt, as a tool for developers to track how design decisions impact monitorability. While the exact prompt we share is still a preliminary version under ongoing development, we are sharing it now in the hopes that others in the community will find it useful. Our method helps measure the default monitorability of CoT - it should be seen as a complement, not a replacement, for the adversarial stress-testing needed to test robustness against deliberately evasive models.

This paper presents an intelligent and transparent AI-driven system for Credit Risk Assessment using three state-of-the-art ensemble machine learning models combined with Explainable AI (XAI) techniques. The system leverages XGBoost, LightGBM, and Random Forest algorithms for predictive analysis of loan default risks, addressing the challenges of model interpretability using SHAP and LIME. Preprocessing steps include custom imputation, one-hot encoding, and standardization. Class imbalance is managed using SMOTE, and hyperparameter tuning is performed with GridSearchCV. The model is evaluated on multiple performance metrics including ROC-AUC, precision, recall, and F1-score. LightGBM emerges as the most business-optimal model with the highest accuracy and best trade off between approval and default rates. Furthermore, the system generates applicant-specific XAI visual reports and business impact summaries to ensure transparent decision-making.

This work presents xEEGNet, a novel, compact, and explainable neural network for EEG data analysis. It is fully interpretable and reduces overfitting through major parameter reduction. As an applicative use case, we focused on classifying common dementia conditions, Alzheimer's and frontotemporal dementia, versus controls. xEEGNet is broadly applicable to other neurological conditions involving spectral alterations. We initially used ShallowNet, a simple and popular model from the EEGNet-family. Its structure was analyzed and gradually modified to move from a "black box" to a more transparent model, without compromising performance. The learned kernels and weights were examined from a clinical standpoint to assess medical relevance. Model variants, including ShallowNet and the final xEEGNet, were evaluated using robust Nested-Leave-N-Subjects-Out cross-validation for unbiased performance estimates. Variability across data splits was explained using embedded EEG representations, grouped by class and set, with pairwise separability to quantify group distinction. Overfitting was assessed through training-validation loss correlation and training speed. xEEGNet uses only 168 parameters, 200 times fewer than ShallowNet, yet retains interpretability, resists overfitting, achieves comparable median performance (-1.5%), and reduces variability across splits. This variability is explained by embedded EEG representations: higher accuracy correlates with greater separation between test set controls and Alzheimer's cases, without significant influence from training data. xEEGNet's ability to filter specific EEG bands, learn band-specific topographies, and use relevant spectral features demonstrates its interpretability. While large deep learning models are often prioritized for performance, this study shows smaller architectures like xEEGNet can be equally effective in EEG pathology classification.

This paper compares model-agnostic and model-specific approaches to explainable AI (XAI) in deep learning image classification. I examine how LIME and SHAP (model-agnostic methods) differ from Grad-CAM and Guided Backpropagation (model-specific methods) when interpreting ResNet50 predictions across diverse image categories. Through extensive testing with various species from dogs and birds to insects I found that each method reveals different aspects of the models decision-making process. Model-agnostic techniques provide broader feature attribution that works across different architectures, while model-specific approaches excel at highlighting precise activation regions with greater computational efficiency. My analysis shows there is no "one-size-fits-all" solution for model interpretability. Instead, combining multiple XAI methods offers the most comprehensive understanding of complex models particularly valuable in high-stakes domains like healthcare, autonomous vehicles, and financial services where transparency is crucial. This comparative framework provides practical guidance for selecting appropriate interpretability techniques based on specific application needs and computational constraints.