arXiv subjects

Niklas Kühl

Publications and source records attributed to Niklas Kühl.

At least 37 records · Page 2Linked to original sources

arXiv2025

A Survey of AI Reliance

Although artificial intelligence (AI) systems are becoming increasingly indispensable, research into how humans rely on these systems (AI reliance) is lagging behind. To advance this research, this survey presents a novel, comprehensive sociotechnical perspective on AI reliance, essential to fully understand the phenomenon. To address these challenges, the survey introduces a categorization framework resulting in a morphological box, which guides rigorous AI reliance research. Further, the survey identifies the core influences on AI reliance within the components of a sociotechnical system and discusses current limitations alongside emerging future research avenues to form a research agenda.

cs.HC↗

arXiv2025

Incremental Model Order Reduction of Smoothed-Particle Hydrodynamic Simulations

Engineering simulations are usually based on complex, grid-based, or mesh-free methods for solving partial differential equations. The results of these methods cover large fields of physical quantities at very many discrete spatial locations and temporal points. Efficient compression methods can be helpful for processing and reusing such large amounts of data. A compression technique is attractive if it causes only a small additional effort and the loss of information with strong compression is low. The paper presents the development of an incremental Singular Value Decomposition (SVD) strategy for compressing time-dependent particle simulation results. The approach is based on an algorithm that was previously developed for grid-based, regular snapshot data matrices. It is further developed here to process highly irregular data matrices generated by particle simulation methods during simulation. Various aspects important for information loss, computational effort and storage requirements are discussed, and corresponding solution techniques are investigated. These include the development of an adaptive rank truncation approach, the assessment of imputation strategies to close snapshot matrix gaps caused by temporarily inactive particles, a suggestion for sequencing the data history into temporal windows as well as bundling the SVD updates. The simulation-accompanying method is embedded in a parallel, industrialized Smoothed-Particle Hydrodynamics software and applied to several 2D and 3D test cases. The proposed approach reduces the memory requirement by about 90% and increases the computational effort by about 10%, while preserving the required accuracy. For the final application of a water turbine, the temporal evolution of the force and torque values for the compressed and simulated data is in excellent agreement.

physics.flu-dyn↗

arXiv2025

Do Edges Matter? Investigating Edge-Enhanced Pre-Training for Medical Image Segmentation

Medical image segmentation is crucial for disease diagnosis and treatment planning, yet developing robust segmentation models often requires substantial computational resources and large datasets. Existing research shows that pre-trained and finetuned foundation models can boost segmentation performance. However, questions remain about how particular image preprocessing steps may influence segmentation performance across different medical imaging modalities. In particular, edges-abrupt transitions in pixel intensity-are widely acknowledged as vital cues for object boundaries but have not been systematically examined in the pre-training of foundation models. We address this gap by investigating to which extend pre-training with data processed using computationally efficient edge kernels, such as kirsch, can improve cross-modality segmentation capabilities of a foundation model. Two versions of a foundation model are first trained on either raw or edge-enhanced data across multiple medical imaging modalities, then finetuned on selected raw subsets tailored to specific medical modalities. After systematic investigation using the medical domains Dermoscopy, Fundus, Mammography, Microscopy, OCT, US, and XRay, we discover both increased and reduced segmentation performance across modalities using edge-focused pre-training, indicating the need for a selective application of this approach. To guide such selective applications, we propose a meta-learning strategy. It uses standard deviation and image entropy of the raw image to choose between a model pre-trained on edge-enhanced or on raw data for optimal performance. Our experiments show that integrating this meta-learning layer yields an overall segmentation performance improvement across diverse medical imaging tasks by 16.42% compared to models pre-trained on edge-enhanced data only and 19.30% compared to models pre-trained on raw data only.

cs.CV↗

arXiv2025

A Multi-Level Strategy for Deepfake Content Moderation under EU Regulation

The growing availability and use of deepfake technologies increases risks for democratic societies, e.g., for political communication on online platforms. The EU has responded with transparency obligations for providers and deployers of Artificial Intelligence (AI) systems and online platforms. This includes marking deepfakes during generation and labeling deepfakes when they are shared. However, the lack of industry and enforcement standards poses an ongoing challenge. Through a multivocal literature review, we summarize methods for marking, detecting, and labeling deepfakes and assess their effectiveness under EU regulation. Our results indicate that individual methods fail to meet regulatory and practical requirements. Therefore, we propose a multi-level strategy combining the strengths of existing methods. To account for the masses of content on online platforms, our multi-level strategy provides scalability and practicality via a simple scoring mechanism. At the same time, it is agnostic to types of deepfake technology and allows for context-specific risk weighting.

cs.CY↗

arXiv2025

Deepfakes in Criminal Investigations: Interdisciplinary Research Directions for CMC Research

The emergence of deepfake technologies offers both opportunities and significant challenges. While commonly associated with deception, misinformation, and fraud, deepfakes may also enable novel applications in high-stakes contexts such as criminal investigations. However, these applications raise complex technological, ethical, and legal questions. We adopt an interdisciplinary approach, drawing on computer science, philosophy, and law, to examine what it takes to responsibly use deepfakes in criminal investigations and argue that computer-mediated communication (CMC) research, especially based on social media corpora, can provide crucial insights for understanding the potential harms and benefits of deepfakes. Our analysis outlines key research directions for the CMC community and underscores the need for interdisciplinary collaboration in this evolving domain.

cs.CY↗

arXiv2025

It's only fair when I think it's fair: How Gender Bias Alignment Undermines Distributive Fairness in Human-AI Collaboration

Human-AI collaboration is increasingly relevant in consequential areas where AI recommendations support human discretion. However, human-AI teams' effectiveness, capability, and fairness highly depend on human perceptions of AI. Positive fairness perceptions have been shown to foster trust and acceptance of AI recommendations. Yet, work on confirmation bias highlights that humans selectively adhere to AI recommendations that align with their expectations and beliefs -- despite not being necessarily correct or fair. This raises the question whether confirmation bias also transfers to the alignment of gender bias between human and AI decisions. In our study, we examine how gender bias alignment influences fairness perceptions and reliance. The results of a 2x2 between-subject study highlight the connection between gender bias alignment, fairness perceptions, and reliance, demonstrating that merely constructing a ``formally fair'' AI system is insufficient for optimal human-AI collaboration; ultimately, AI recommendations will likely be overridden if biases do not align.

cs.HC↗

arXiv2025

Honey, I Shrunk the Language Model: Impact of Knowledge Distillation Methods on Performance and Explainability

Artificial Intelligence (AI) has increasingly influenced modern society, recently in particular through significant advancements in Large Language Models (LLMs). However, high computational and storage demands of LLMs still limit their deployment in resource-constrained environments. Knowledge distillation addresses this challenge by training a small student model from a larger teacher model. Previous research has introduced several distillation methods for both generating training data and for training the student model. Despite their relevance, the effects of state-of-the-art distillation methods on model performance and explainability have not been thoroughly investigated and compared. In this work, we enlarge the set of available methods by applying critique-revision prompting to distillation for data generation and by synthesizing existing methods for training. For these methods, we provide a systematic comparison based on the widely used Commonsense Question-Answering (CQA) dataset. While we measure performance via student model accuracy, we employ a human-grounded study to evaluate explainability. We contribute new distillation methods and their comparison in terms of both performance and explainability. This should further advance the distillation of small language models and, thus, contribute to broader applicability and faster diffusion of LLM technology.

cs.CL↗

arXiv2025

Differential Equation Based Wall Distance Approaches for Maritime Engineering Flows

The paper is concerned with modeling and simulating approaches of wall distance functions based on Partial Differential Equations (PDE). The distance to the nearest wall is required for many industrial problems in Computational Fluid Dynamics (CFD). The first part of the manuscript addresses fundamental aspects of wall distance modeling and simulation. The following approaches are considered: Nonlinear and linear p-Poisson and Screened-Poisson methods, Eikonal and regularized Eikonal or Hamilton-Jacobi methods, and alternatives using Laplace equations. Following the definition of boundary and initial conditions, the discrete approximation and relevant measures to increase its numerical robustness are described. In the second part, the different methods are applied to hydrodynamic and aerodynamic flow applications from maritime engineering, each relying on Shear Stress Transport (SST) strategies for turbulence modeling that require the distance to the nearest wall at different procedural points. The hydrodynamic behavior of a model scale bulk carrier cruising at ReL=7.246E+6 and Fn = 0.142 is investigated on the influence of the wall distance formulation for predicting resistance and propulsion behavior in conjunction with statistical turbulence modeling method. It is shown that the different wall distance modeling barely influences relevant integral hydrodynamic quantities such as drag, trim, and sinkage, and related errors are in the range of O(0.1%) and, therefore, significantly below typical modeling, discretization, and approximation errors. Subsequently, the wall distance methods were investigated for the aerodynamic analysis of a full-scale feeder ship at ReL = 5E+08. A hybrid averaged/filtered approach, in line with the Improved Delayed Detached Eddy Simulation (IDDES) model, is utilized, and the results indicate an improved sensitivity to the choice of the wall distance model.

physics.flu-dyn↗

arXiv2025

Beware of "Explanations" of AI

Understanding the decisions made and actions taken by increasingly complex AI system remains a key challenge. This has led to an expanding field of research in explainable artificial intelligence (XAI), highlighting the potential of explanations to enhance trust, support adoption, and meet regulatory standards. However, the question of what constitutes a "good" explanation is dependent on the goals, stakeholders, and context. At a high level, psychological insights such as the concept of mental model alignment can offer guidance, but success in practice is challenging due to social and technical factors. As a result of this ill-defined nature of the problem, explanations can be of poor quality (e.g. unfaithful, irrelevant, or incoherent), potentially leading to substantial risks. Instead of fostering trust and safety, poorly designed explanations can actually cause harm, including wrong decisions, privacy violations, manipulation, and even reduced AI adoption. Therefore, we caution stakeholders to beware of explanations of AI: while they can be vital, they are not automatically a remedy for transparency or responsible AI adoption, and their misuse or limitations can exacerbate harm. Attention to these caveats can help guide future research to improve the quality and impact of AI explanations.

cs.LG↗

arXiv2025

Towards Human-Understandable Multi-Dimensional Concept Discovery

Concept-based eXplainable AI (C-XAI) aims to overcome the limitations of traditional saliency maps by converting pixels into human-understandable concepts that are consistent across an entire dataset. A crucial aspect of C-XAI is completeness, which measures how well a set of concepts explains a model's decisions. Among C-XAI methods, Multi-Dimensional Concept Discovery (MCD) effectively improves completeness by breaking down the CNN latent space into distinct and interpretable concept subspaces. However, MCD's explanations can be difficult for humans to understand, raising concerns about their practical utility. To address this, we propose Human-Understandable Multi-dimensional Concept Discovery (HU-MCD). HU-MCD uses the Segment Anything Model for concept identification and implements a CNN-specific input masking technique to reduce noise introduced by traditional masking methods. These changes to MCD, paired with the completeness relation, enable HU-MCD to enhance concept understandability while maintaining explanation faithfulness. Our experiments, including human subject studies, show that HU-MCD provides more precise and reliable explanations than existing C-XAI methods. The code is available at https://github.com/grobruegge/hu-mcd.

cs.CV↗

arXiv2025

Human Preferences in Large Language Model Latent Space: A Technical Analysis on the Reliability of Synthetic Data in Voting Outcome Prediction

Generative AI (GenAI) is increasingly used in survey contexts to simulate human preferences. While many research endeavors evaluate the quality of synthetic GenAI data by comparing model-generated responses to gold-standard survey results, fundamental questions about the validity and reliability of using LLMs as substitutes for human respondents remain. Our study provides a technical analysis of how demographic attributes and prompt variations influence latent opinion mappings in large language models (LLMs) and evaluates their suitability for survey-based predictions. Using 14 different models, we find that LLM-generated data fails to replicate the variance observed in real-world human responses, particularly across demographic subgroups. In the political space, persona-to-party mappings exhibit limited differentiation, resulting in synthetic data that lacks the nuanced distribution of opinions found in survey data. Moreover, we show that prompt sensitivity can significantly alter outputs for some models, further undermining the stability and predictiveness of LLM-based simulations. As a key contribution, we adapt a probe-based methodology that reveals how LLMs encode political affiliations in their latent space, exposing the systematic distortions introduced by these models. Our findings highlight critical limitations in AI-generated survey data, urging caution in its use for public opinion research, social science experimentation, and computational behavioral modeling.

cs.LG↗

arXiv2025

Elliptic Relaxation Strategies to Support Numerical Stability of Segregated Continuous Adjoint Flow Solvers

This paper introduces a novel method for numerically stabilizing sequential continuous adjoint flow solvers utilizing an elliptic relaxation strategy. The proposed approach is formulated as a Partial Differential Equation (PDE) containing a single user-defined parameter, which analytical investigations reveal to represent the filter width of a probabilistic density function or Gaussian kernel. Key properties of the approach include (a) smoothing features with redistribution capabilities while (b) preserving integral properties. The technique targets explicit adjoint cross-coupling terms, such as the Adjoint Transpose Convection (ATC) term, which frequently causes numerical instabilities, especially on unstructured grids common in industrial applications. A trade-off is made by sacrificing sensitivity consistency to achieve enhanced numerical robustness. The method is validated on a two-phase, laminar, two-dimensional cylinder flow test case at Re=20 and Fn=0.75, focusing on minimizing resistance or maximizing lift. A range of homogeneous and inhomogeneous filter widths is evaluated. Subsequently, the relaxation method is employed to stabilize adjoint simulations during shape optimizations that aim at drag reduction of ship hulls. Two case studies are considered: A model-scale bulk carrier traveling at Re=7.246E+06 and Fn=0.142 as well as a harbor ferry cruising at Re=2.43E+08 and Fn=0.4 in full-scale conditions. Both cases, characterized by unstructured grids prone to adjoint divergence, demonstrate the effectiveness of the proposed method in overcoming stability challenges. The resulting optimizations achieve superior outcomes compared to approaches that omit problematic coupling terms.

physics.flu-dyn↗

arXiv2025

Human Delegation Behavior in Human-AI Collaboration: The Effect of Contextual Information

The integration of artificial intelligence (AI) into human decision-making processes at the workplace presents both opportunities and challenges. One promising approach to leverage existing complementary capabilities is allowing humans to delegate individual instances of decision tasks to AI. However, enabling humans to delegate instances effectively requires them to assess several factors. One key factor is the analysis of both their own capabilities and those of the AI in the context of the given task. In this work, we conduct a behavioral study to explore the effects of providing contextual information to support this delegation decision. Specifically, we investigate how contextual information about the AI and the task domain influence humans' delegation decisions to an AI and their impact on the human-AI team performance. Our findings reveal that access to contextual information significantly improves human-AI team performance in delegation settings. Finally, we show that the delegation behavior changes with the different types of contextual information. Overall, this research advances the understanding of computer-supported, collaborative work and provides actionable insights for designing more effective collaborative systems.

cs.HC↗

arXiv2025

Don't be Fooled: The Misinformation Effect of Explanations in Human-AI Collaboration

Across various applications, humans increasingly use black-box artificial intelligence (AI) systems without insight into these systems' reasoning. To counter this opacity, explainable AI (XAI) methods promise enhanced transparency and interpretability. While recent studies have explored how XAI affects human-AI collaboration, few have examined the potential pitfalls caused by incorrect explanations. The implications for humans can be far-reaching but have not been explored extensively. To investigate this, we ran a study (n=160) on AI-assisted decision-making in which humans were supported by XAI. Our findings reveal a misinformation effect when incorrect explanations accompany correct AI advice with implications post-collaboration. This effect causes humans to infer flawed reasoning strategies, hindering task execution and demonstrating impaired procedural knowledge. Additionally, incorrect explanations compromise human-AI team-performance during collaboration. With our work, we contribute to HCI by providing empirical evidence for the negative consequences of incorrect explanations on humans post-collaboration and outlining guidelines for designers of AI.

cs.HC↗

arXiv2025

Towards a Problem-Oriented Domain Adaptation Framework for Machine Learning

Domain adaptation is a sub-field of machine learning that involves transferring knowledge from a source domain to perform the same task in the target domain. It is a typical challenge in machine learning that arises, e.g., when data is obtained from various sources or when using a data basis that changes over time. Recent advances in the field offer promising methods, but it is still challenging for researchers and practitioners to determine if domain adaptation is suitable for a given problem -- and, subsequently, to select the appropriate approach. This article employs design science research to develop a problem-oriented framework for domain adaptation, which is matured in three evaluation episodes. We describe a framework that distinguishes between five domain adaptation scenarios, provides recommendations for addressing each scenario, and offers guidelines for determining if a problem falls into one of these scenarios. During the multiple evaluation episodes, the framework is tested on artificial and real-world datasets and an experimental study involving 100 participants. The evaluation demonstrates that the framework has the explanatory power to capture any domain adaptation problem effectively. In summary, we provide clear guidance for researchers and practitioners who want to employ domain adaptation but lack in-depth knowledge of the possibilities.

cs.LG↗

arXiv2024

Complementarity in Human-AI Collaboration: Concept, Sources, and Evidence

Artificial intelligence (AI) has the potential to significantly enhance human performance across various domains. Ideally, collaboration between humans and AI should result in complementary team performance (CTP) -- a level of performance that neither of them can attain individually. So far, however, CTP has rarely been observed, suggesting an insufficient understanding of the principle and the application of complementarity. Therefore, we develop a general concept of complementarity and formalize its theoretical potential as well as the actual realized effect in decision-making situations. Moreover, we identify information and capability asymmetry as the two key sources of complementarity. Finally, we illustrate the impact of each source on complementarity potential and effect in two empirical studies. Our work provides researchers with a comprehensive theoretical foundation of human-AI complementarity in decision-making and demonstrates that leveraging these sources constitutes a viable pathway towards designing effective human-AI collaboration, i.e., the realization of CTP.

cs.HC↗

arXiv2024

A Multivocal Literature Review on Privacy and Fairness in Federated Learning

Federated Learning presents a way to revolutionize AI applications by eliminating the necessity for data sharing. Yet, research has shown that information can still be extracted during training, making additional privacy-preserving measures such as differential privacy imperative. To implement real-world federated learning applications, fairness, ranging from a fair distribution of performance to non-discriminative behaviour, must be considered. Particularly in high-risk applications (e.g. healthcare), avoiding the repetition of past discriminatory errors is paramount. As recent research has demonstrated an inherent tension between privacy and fairness, we conduct a multivocal literature review to examine the current methods to integrate privacy and fairness in federated learning. Our analyses illustrate that the relationship between privacy and fairness has been neglected, posing a critical risk for real-world applications. We highlight the need to explore the relationship between privacy, fairness, and performance, advocating for the creation of integrated federated learning frameworks.

cs.LG↗

arXiv2024

Utilizing Data Fingerprints for Privacy-Preserving Algorithm Selection in Time Series Classification: Performance and Uncertainty Estimation on Unseen Datasets

The selection of algorithms is a crucial step in designing AI services for real-world time series classification use cases. Traditional methods such as neural architecture search, automated machine learning, combined algorithm selection, and hyperparameter optimizations are effective but require considerable computational resources and necessitate access to all data points to run their optimizations. In this work, we introduce a novel data fingerprint that describes any time series classification dataset in a privacy-preserving manner and provides insight into the algorithm selection problem without requiring training on the (unseen) dataset. By decomposing the multi-target regression problem, only our data fingerprints are used to estimate algorithm performance and uncertainty in a scalable and adaptable manner. Our approach is evaluated on the 112 University of California riverside benchmark datasets, demonstrating its effectiveness in predicting the performance of 35 state-of-the-art algorithms and providing valuable insights for effective algorithm selection in time series classification service systems, improving a naive baseline by 7.32% on average in estimating the mean performance and 15.81% in estimating the uncertainty.

cs.LG↗