Research Papers

Large language models often fail at logical reasoning when semantic heuristics conflict with decisive evidence - a phenomenon we term cognitive traps. To address this fundamental limitation, we introduce the Deliberative Reasoning Network (DRN), a novel paradigm that reframes logical reasoning from probability maximization to uncertainty minimization. Instead of asking "Which answer is most likely?", DRN asks "Which hypothesis has the most internally consistent evidence?". DRN achieves intrinsic interpretability by explicitly tracking belief states and quantifying epistemic uncertainty for competing hypotheses through an iterative evidence synthesis process. We validate our approach through two complementary architectures - a bespoke discriminative model that embodies the core uncertainty minimization principle, and a lightweight verification module that enhances existing generative LLMs. Evaluated on LCR-1000, our new adversarial reasoning benchmark designed to expose cognitive traps, the bespoke DRN achieves up to 15.2% improvement over standard baselines. When integrated as a parameter-efficient verifier with Mistral-7B, our hybrid system boosts accuracy from 20% to 80% on the most challenging problems. Critically, DRN demonstrates strong zero-shot generalization, improving TruthfulQA performance by 23.6% without additional training, indicating that uncertainty-driven deliberation learns transferable reasoning principles. We position DRN as a foundational, verifiable System 2 reasoning component for building more trustworthy AI systems.

Fast and Accurate Explanations of Distance-Based Classifiers by Uncovering Latent Explanatory Structures

Florian Bley, Jacob Kauffmann, Simon León Krug +2

Distance-based classifiers, such as k-nearest neighbors and support vector machines, continue to be a workhorse of machine learning, widely used in science and industry. In practice, to derive insights from these models, it is also important to ensure that their predictions are explainable. While the field of Explainable AI has supplied methods that are in principle applicable to any model, it has also emphasized the usefulness of latent structures (e.g. the sequence of layers in a neural network) to produce explanations. In this paper, we contribute by uncovering a hidden neural network structure in distance-based classifiers (consisting of linear detection units combined with nonlinear pooling layers) upon which Explainable AI techniques such as layer-wise relevance propagation (LRP) become applicable. Through quantitative evaluations, we demonstrate the advantage of our novel explanation approach over several baselines. We also show the overall usefulness of explaining distance-based models through two practical use cases.

Accurate estimation of postmenstrual age (PMA) at scan is crucial for assessing neonatal development and health. While deep learning models have achieved high accuracy in predicting PMA from brain MRI, they often function as black boxes, offering limited transparency and interpretability in clinical decision support. In this work, we address the dual challenge of accuracy and interpretability by adapting a multimodal large language model (MLLM) to perform both precise PMA prediction and clinically relevant explanation generation. We introduce a parameter-efficient fine-tuning (PEFT) strategy using instruction tuning and Low-Rank Adaptation (LoRA) applied to the Qwen2.5-VL-7B model. The model is trained on four 2D cortical surface projection maps derived from neonatal MRI scans. By employing distinct prompts for training and inference, our approach enables the MLLM to handle a regression task during training and generate clinically relevant explanations during inference. The fine-tuned model achieves a low prediction error with a 95 percent confidence interval of 0.78 to 1.52 weeks, while producing interpretable outputs grounded in developmental features, marking a significant step toward transparent and trustworthy AI systems in perinatal neuroscience.

With the development of generative artificial intelligence (GenAI) tools to create art, stakeholders cannot come to an agreement on the value of these works. In this study we uncovered the mixed opinions surrounding art made by AI. We developed two versions of a dance performance augmented by technology either with or without GenAI. For each version we informed audiences of the performance's development either before or after a survey on their perceptions of the performance. There were thirty-nine participants (13 males, 26 female) divided between the four performances. Results demonstrated that individuals were more inclined to attribute artistic merit to works made by GenAI when they were unaware of its use. We present this case study as a call to address the importance of utilizing the social context and the users' interpretations of GenAI in shaping a technical explanation, leading to a greater discussion that can bridge gaps in understanding.

Explanations are crucial for building trustworthy AI systems, but a gap often exists between the explanations provided by models and those needed by users. To address this gap, we introduce MetaExplainer, a neuro-symbolic framework designed to generate user-centered explanations. Our approach employs a three-stage process: first, we decompose user questions into machine-readable formats using state-of-the-art large language models (LLM); second, we delegate the task of generating system recommendations to model explainer methods; and finally, we synthesize natural language explanations that summarize the explainer outputs. Throughout this process, we utilize an Explanation Ontology to guide the language models and explainer methods. By leveraging LLMs and a structured approach to explanation generation, MetaExplainer aims to enhance the interpretability and trustworthiness of AI systems across various applications, providing users with tailored, question-driven explanations that better meet their needs. Comprehensive evaluations of MetaExplainer demonstrate a step towards evaluating and utilizing current state-of-the-art explanation frameworks. Our results show high performance across all stages, with a 59.06% F1-score in question reframing, 70% faithfulness in model explanations, and 67% context-utilization in natural language synthesis. User studies corroborate these findings, highlighting the creativity and comprehensiveness of generated explanations. Tested on the Diabetes (PIMA Indian) tabular dataset, MetaExplainer supports diverse explanation types, including Contrastive, Counterfactual, Rationale, Case-Based, and Data explanations. The framework's versatility and traceability from using ontology to guide LLMs suggest broad applicability beyond the tested scenarios, positioning MetaExplainer as a promising tool for enhancing AI explainability across various domains.

Obtaining human-readable symbolic formulas via genetic programming-based symbolic distillation of a deep neural network trained on the target dataset presents a promising yet underexplored path towards explainable artificial intelligence (XAI); however, the standard pipeline frequently yields symbolic models with poor predictive accuracy. We identify a fundamental misalignment in functional complexity as the primary barrier to achieving better accuracy: standard Artificial Neural Networks (ANNs) often learn accurate but highly irregular functions, while Symbolic Regression typically prioritizes parsimony, often resulting in a much simpler class of models that are unable to sufficiently distill or learn from the ANN teacher. To bridge this gap, we propose a framework that actively regularizes the teacher's functional smoothness using Jacobian and Lipschitz penalties, aiming to distill better student models than the standard pipeline. We characterize the trade-off between predictive accuracy and functional complexity through a robust study involving 20 datasets and 50 independent trials. Our results demonstrate that students distilled from smoothness-regularized teachers achieve statistically significant improvements in R^2 scores, compared to the standard pipeline. We also perform ablation studies on the student model algorithm. Our findings suggest that smoothness alignment between teacher and student models is a critical factor for symbolic distillation.

Leveraging recent advances in the analysis of synergy and redundancy in systems of random variables, an adaptive version of the widely used metric Leave One Covariate Out (LOCO) has been recently proposed to quantify cooperative effects in feature importance (Hi-Fi), a key technique in explainable artificial intelligence (XAI), so as to disentangle high-order effects involving a particular input feature in regression problems. Differently from standard feature importance tools, where a single score measures the relevance of each feature, each feature is here characterized by three scores, a two-body (unique) score and higher-order scores (redundant and synergistic). This paper presents a framework to assign those three scores (unique, redundant, and synergistic) to each individual pattern of the data set, while comparing it with the well-known measure of feature importance named {\it Shapley effect}. To illustrate the potential of the proposed framework, we focus on a One-Health application: the relation between air pollutants and Alzheimer's disease mortality rate. Our main result is the synergistic association between features related to $O_3$ and $NO_2$ with mortality, especially in the provinces of Bergamo e Brescia; notably also the density of urban green areas displays synergistic influence with pollutants for the prediction of AD mortality. Our results place local Hi-Fi as a promising tool of wide applicability, which opens new perspectives for XAI as well as to analyze high-order relationships in complex systems.

In recent years, Explainable AI has gained traction as an approach to enhancing model interpretability and transparency, particularly in complex models such as detection transformers. Despite rapid advancements, a substantial research gap remains in understanding the distinct roles of internal components - knowledge that is essential for improving transparency and efficiency. Inspired by neuroscientific ablation studies, which investigate the functions of brain regions through selective impairment, we systematically analyze the impact of ablating key components in three state-of-the-art detection transformer models: Detection transformer (DETR), deformable detection transformer (DDETR), and DETR with improved denoising anchor boxes (DINO). The ablations target query embeddings, encoder and decoder multi-head self-attentions (MHSA) as well as decoder multi-head cross-attention (MHCA) layers. We evaluate the effects of these ablations on the performance metrics gIoU and F1-score, quantifying effects on both the classification and regression sub-tasks on the COCO dataset. To facilitate reproducibility and future research, we publicly release the DeepDissect library. Our findings reveal model-specific resilience patterns: while DETR is particularly sensitive to ablations in encoder MHSA and decoder MHCA, DDETR's multi-scale deformable attention enhances robustness, and DINO exhibits the greatest resilience due to its look-forward twice update rule, which helps distributing knowledge across blocks. These insights also expose structural redundancies, particularly in DDETR's and DINO's decoder MHCA layers, highlighting opportunities for model simplification without sacrificing performance. This study advances XAI for DETRs by clarifying the contributions of internal components to model performance, offering insights to optimize and improve transparency and efficiency in critical applications.

Ensuring the trustworthiness and robustness of deep learning models remains a fundamental challenge, particularly in high-stakes scientific applications. In this study, we present a framework called attention-guided training that combines explainable artificial intelligence techniques with quantitative evaluation and domain-specific priors to guide model attention. We demonstrate that domain-specific feedback on model explanations during training can enhance the model's generalization capabilities. We validate our approach on the task of semantic crack tip segmentation in digital image correlation data, which is a key application in the fracture mechanical characterization of materials. By aligning model attention with physically meaningful stress fields, such as those described by Williams' analytical solution, attention-guided training ensures that the model focuses on physically relevant regions. This finally leads to improved generalization and more faithful explanations.

Prostate cancer, the second most prevalent male malignancy, requires advanced diagnostic tools. We propose an explainable AI system combining BERT (for textual clinical notes) and Random Forest (for numerical lab data) through a novel multimodal fusion strategy, achieving superior classification performance on PLCO-NIH dataset (98% accuracy, 99% AUC). While multimodal fusion is established, our work demonstrates that a simple yet interpretable BERT+RF pipeline delivers clinically significant improvements - particularly for intermediate cancer stages (Class 2/3 recall: 0.900 combined vs 0.824 numerical/0.725 textual). SHAP analysis provides transparent feature importance rankings, while ablation studies prove textual features' complementary value. This accessible approach offers hospitals a balance of high performance (F1=89%), computational efficiency, and clinical interpretability - addressing critical needs in prostate cancer diagnostics.

In recent years, social media platforms have become prominent spaces for individuals to express their opinions on ongoing events, including criminal incidents. As a result, public sentiment can shift dynamically over time. This study investigates the evolving public perception of crime-related news by classifying user-generated comments into three categories: positive, negative, and neutral. A newly curated dataset comprising 28,528 Bangla-language social media comments was developed for this purpose. We propose a transformer-based model utilizing the XLM-RoBERTa Base architecture, which achieves a classification accuracy of 97%, outperforming existing state-of-the-art methods in Bangla sentiment analysis. To enhance model interpretability, explainable AI technique is employed to identify the most influential features driving sentiment classification. The results underscore the effectiveness of transformer-based models in processing low-resource languages such as Bengali and demonstrate their potential to extract actionable insights that can support public policy formulation and crime prevention strategies.

Consumption of YouTube news videos significantly shapes public opinion and political narratives. While prior works have studied the longitudinal dissemination dynamics of YouTube News videos across extended periods, limited attention has been paid to the short-term trends. In this paper, we investigate the early-stage diffusion patterns and dispersion rate of news videos on YouTube, focusing on the first 24 hours. To this end, we introduce and analyze a rich dataset of over 50,000 videos across 75 countries and six continents. We provide the first quantitative evaluation of the 24-hour half-life of YouTube news videos as well as identify their distinct diffusion patterns. According to the findings, the average 24-hour half-life is approximately 7 hours, with substantial variance both within and across countries, ranging from as short as 2 hours to as long as 15 hours. Additionally, we explore the problem of predicting the latency of news videos' 24-hour half-lives. Leveraging the presented datasets, we train and contrast the performance of 6 different models based on statistical as well as Deep Learning techniques. The difference in prediction results across the models is traced and analyzed. Lastly, we investigate the importance of video- and channel-related predictors through Explainable AI (XAI) techniques. The dataset, analysis codebase and the trained models are released at http://bit.ly/3ILvTLU to facilitate further research in this area.