Research Papers

The field of Artificial Intelligence (AI) and, in particular, the Machine Learning area, counts on a wide range of performance metrics and benchmark data sets to assess the problem-solving effectiveness of its solutions. However, the appearance of research centres, projects or institutions addressing AI solutions from a multidisciplinary and multi-stakeholder perspective suggests a new approach to assessment comprising ethical guidelines, reports or tools and frameworks to help both academia and business to move towards a responsible conceptualisation of AI. They all highlight the relevance of three key aspects: (i) enhancing cooperation among the different stakeholders involved in the design, deployment and use of AI; (ii) promoting multidisciplinary dialogue, including different domains of expertise in this process; and (iii) fostering public engagement to maximise a trusted relation with new technologies and practitioners. In this paper, we introduce the Observatory on Society and Artificial Intelligence (OSAI), an initiative grew out of the project AI4EU aimed at stimulating reflection on a broad spectrum of issues of AI (ethical, legal, social, economic and cultural). In particular, we describe our work in progress around OSAI and suggest how this and similar initiatives can promote a wider appraisal of progress in AI. This will give us the opportunity to present our vision and our modus operandi to enhance the implementation of these three fundamental dimensions.

While decision makers have begun to employ machine learning, machine learning models may make predictions that bias against certain demographic groups. Semi-automated bias detection tools often present reports of automatically-detected biases using a recommendation list or visual cues. However, there is a lack of guidance concerning which presentation style to use in what scenarios. We conducted a small lab study with 16 participants to investigate how presentation style might affect user behaviors in reviewing bias reports. Participants used both a prototype with a recommendation list and a prototype with visual cues for bias detection. We found that participants often wanted to investigate the performance measures that were not automatically detected as biases. Yet, when using the prototype with a recommendation list, they tended to give less consideration to such measures. Grounded in the findings, we propose information load and comprehensiveness as two axes for characterizing bias detection tasks and illustrate how the two axes could be adopted to reason about when to use a recommendation list or visual cues.

To achieve optimal human-system integration in the context of user-AI interaction it is important that users develop a valid representation of how AI works. In most of the everyday interaction with technical systems users construct mental models (i.e., an abstraction of the anticipated mechanisms a system uses to perform a given task). If no explicit explanations are provided by a system (e.g. by a self-explaining AI) or other sources (e.g. an instructor), the mental model is typically formed based on experiences, i.e. the observations of the user during the interaction. The congruence of this mental model and the actual systems functioning is vital, as it is used for assumptions, predictions and consequently for decisions regarding system use. A key question for human-centered AI research is therefore how to validly survey users' mental models. The objective of the present research is to identify suitable elicitation methods for mental model analysis. We evaluated whether mental models are suitable as an empirical research method. Additionally, methods of cognitive tutoring are integrated. We propose an exemplary method to evaluate explainable AI approaches in a human-centered way.

Despite the recent successes of deep learning, such models are still far from some human abilities like learning from few examples, reasoning and explaining decisions. In this paper, we focus on organ annotation in medical images and we introduce a reasoning framework that is based on learning fuzzy relations on a small dataset for generating explanations. Given a catalogue of relations, it efficiently induces the most relevant relations and combines them for building constraints in order to both solve the organ annotation task and generate explanations. We test our approach on a publicly available dataset of medical images where several organs are already segmented. A demonstration of our model is proposed with an example of explained annotations. It was trained on a small training set containing as few as a couple of examples.

The debate on AI ethics largely focuses on technical improvements and stronger regulation to prevent accidents or misuse of AI, with solutions relying on holding individual actors accountable for responsible AI development. While useful and necessary, we argue that this "agency" approach disregards more indirect and complex risks resulting from AI's interaction with the socio-economic and political context. This paper calls for a "structural" approach to assessing AI's effects in order to understand and prevent such systemic risks where no individual can be held accountable for the broader negative impacts. This is particularly relevant for AI applied to systemic issues such as climate change and food security which require political solutions and global cooperation. To properly address the wide range of AI risks and ensure 'AI for social good', agency-focused policies must be complemented by policies informed by a structural approach.

The cyberspace and development of intelligent systems using Artificial Intelligence (AI) creates new challenges to computer professionals, data scientists, regulators and policy makers. For example, self-driving cars raise new technical, ethical, legal and public policy issues. This paper proposes a course named Computers, Ethics, Law, and Public Policy, and suggests a curriculum for such a course. This paper presents ethical, legal, and public policy issues relevant to building and using intelligent systems.

Visualizing the decision-making process in deep neural decision forest

Shichao Li, Kwang-Ting Cheng

Deep neural decision forest (NDF) achieved remarkable performance on various vision tasks via combining decision tree and deep representation learning. In this work, we first trace the decision-making process of this model and visualize saliency maps to understand which portion of the input influence it more for both classification and regression problems. We then apply NDF on a multi-task coordinate regression problem and demonstrate the distribution of routing probabilities, which is vital for interpreting NDF yet not shown for regression problems. The pre-trained model and code for visualization will be available at https://github.com/Nicholasli1995/VisualizingNDF

In Random Forests, proximity distances are a metric representation of data into decision space. By observing how changes in input map to the movement of instances in this space we are able to determine the independent contribution of each feature to the decision-making process. For binary feature vectors, this process is fully specified. As these changes in input move particular instances nearer to the in-group or out-group, the independent contribution of each feature can be uncovered. Using this technique, we are able to calculate the contribution of each feature in determining how black-box decisions were made. This allows explication of the decision-making process, audit of the classifier, and post-hoc analysis of errors in classification.

Ensuring trust in AI systems is essential for the safe and ethical integration of machine learning systems into high-stakes domains such as digital health. Key dimensions, including robustness, explainability, fairness, accountability, and privacy, need to be addressed throughout the AI lifecycle, from problem formulation and data collection to model deployment and human interaction. While various contributions address different aspects of trustworthy AI, a focused synthesis on robustness and explainability, especially tailored to the healthcare context, remains limited. This review addresses that need by organizing recent advancements into an accessible framework, highlighting both technical and practical considerations. We present a structured overview of methods, challenges, and solutions, aiming to support researchers and practitioners in developing reliable and explainable AI solutions for digital health. This review article is organized into three main parts. First, we introduce the pillars of trustworthy AI and discuss the technical and ethical challenges, particularly in the context of digital health. Second, we explore application-specific trust considerations across domains such as intensive care, neonatal health, and metabolic health, highlighting how robustness and explainability support trust. Lastly, we present recent advancements in techniques aimed at improving robustness under data scarcity and distributional shifts, as well as explainable AI methods ranging from feature attribution to gradient-based interpretations and counterfactual explanations. This paper is further enriched with detailed discussions of the contributions toward robustness and explainability in digital health, the development of trustworthy AI systems in the era of LLMs, and various evaluation metrics for measuring trust and related parameters such as validity, fidelity, and diversity.

Not All LLM Reasoning is Visible in the Chain-of-Thought

Vatsal Baherwani, Tom Goldstein, Ashwinee Panda

A key question for AI safety is whether a language model expresses all of its reasoning in its output tokens. We demonstrate a concrete failure mode where frontier models exhibit invisible reasoning by leveraging semantically irrelevant filler tokens to improve performance on synthetic reasoning tasks. We evaluate 13 frontier language models across three tasks and find that many models benefit significantly from filler tokens, with accuracy improvements of up to 13 percentage points. The benefit depends on which tokens are used and differs across models. We further show that filler tokens enable Claude Opus 4.5 to satisfy a hidden modular arithmetic constraint without sacrificing accuracy on its primary task, demonstrating that invisible reasoning can serve objectives entirely invisible to CoT monitoring. Reinforcement learning gives Qwen3-235B strong preferences over filler token content, but neither RL nor supervised fine-tuning produces a filler token benefit that persists at test time. Our results indicate that frontier models already perform consequential computation with no interpretable trace in their output tokens.

Advancements in data-driven machine learning have emerged as a pivotal element in supporting automotive software systems (ASSs) engineering across various levels of the V-development process. Duringsystemverificationandvalidation,theintegrationofanintelligent fault detection anddiagnosis (FDD) model with test recordings analysis process serves as a powerful tool for efficiency ensuring functional safety. However, the lack of interpretability of the black-box FDD models developed not only hinders understanding of the cause underlying the prediction, but also prevents the model from being adapted based on the prediction result. This, in turn, increases the computational cost required for developingacomplexFDDmodelandlimitsconfidenceinreal-timesafety-criticalapplications.To address this challenge, a novel explainable method for fault detection, identification, and localization is proposed in this article with the aim of providing a clear understanding of the logic behind the prediction outcome. To this end, a hybrid 1dCNN-GRU-based intelligent model was developed to analyze the recordings from the real-time validation process of ASSs. The employment of explainable AI techniques, i.e., IGs, DeepLIFT, Gradient SHAP, and DeepLIFT SHAP, was instrumental in enabling model adaptation and facilitating the root cause analysis (RCA). The proposed approach is applied to the real time dataset collected during a virtual test drive performed by the user on hardware in the loop system.

Large language models increasingly function as epistemic agents -- entities that can 1) autonomously pursue epistemic goals and 2) actively shape our shared knowledge environment. They curate the information we receive, often supplanting traditional search-based methods, and are frequently used to generate both personal and deeply specialized advice. How they perform these functions, including whether they are reliable and properly calibrated to both individual and collective epistemic norms, is therefore highly consequential for the choices we make. We argue that the potential impact of epistemic AI agents on practices of knowledge creation, curation and synthesis, particularly in the context of complex multi-agent interactions, creates new informational interdependencies that necessitate a fundamental shift in evaluation and governance of AI. While a well-calibrated ecosystem could augment human judgment and collective decision-making, poorly aligned agents risk causing cognitive deskilling and epistemic drift, making the calibration of these models to human norms a high-stakes necessity. To ensure a beneficial human-AI knowledge ecosystem, we propose a framework centered on building and cultivating the trustworthiness of epistemic AI agents; aligning AI these agents with human epistemic goals; and reinforcing the surrounding socio-epistemic infrastructure. In this context, trustworthy AI agents must demonstrate epistemic competence, robust falsifiability, and epistemically virtuous behaviors, supported by technical provenance systems and "knowledge sanctuaries" designed to protect human resilience. This normative roadmap provides a path toward ensuring that future AI systems act as reliable partners in a robust and inclusive knowledge ecosystem.