arXiv subjects

Katja Hose

Publications and source records attributed to Katja Hose.

At least 19 recordsLinked to original sources

arXiv2026

Value-Preserving Architectures for Agentic AI Systems

The emergence of agentic AI and LLM-based multi-agent systems (MAS) presents unprecedented opportunities for automating complex tasks, while simultaneously raising critical concerns about the preservation of fundamental human-centered values, such as privacy, fairness, and safety. Although software engineering has traditionally focused on functional correctness, the adoption of LLMs and AI agents into complex socio-technical systems has intensified the need for responsible software engineering and robust value alignment. In MAS, architectural design decisions, such as coordination mechanisms, communication protocols, and system topologies, play a central role in shaping system behavior and the outcomes they produce. This paper argues that architectural choices influence not only the functionality and performance of MAS but can also promote value-oriented system behavior. Therefore, we investigate how different architectural designs support different human-centered values, discussing the following value-preserving architectural patterns: (i) a privacy-aware architecture with a federated topology, (ii) a distributed architecture to promote pluralism and diversity, and (iii) a guard-agent architecture to detect and mitigate unfairness. Finally, we introduce representative use cases to illustrate the proposed architectures in real-world scenarios. By linking architectural design with human-centered values, this work lays the foundation for a unified set of architectural patterns and guidelines towards the design of trustworthy MAS.

cs.AI↗

arXiv2026

DataKernelBench: Can LLMs Optimize Database Queries on GPUs?

GPUs increasingly accelerate database systems, but query-specific peak performance still often relies on hand-written kernels. Existing LLM kernel benchmarks focus on machine learning operators, leaving irregular, heterogeneous, data-movement-heavy database-style operators untested. We introduce DataKernelBench, which translates SQL into validated PyTorch TorchPlan programs and evaluates LLMs that optimize either the core tensor-bounded snippet or the full query in CUDA or Triton through execution-guided repair. Across ten proprietary and open-weight models on TPC-H SF10 with an H100 GPU, the strongest full-query CUDA configuration achieves $2.11\times$ speedup over the TorchPlan baseline at full pass rate. We find that higher-performing implementations commonly use kernel fusion and execution-strategy changes, stronger models benefit most from full-query specialization, and workload context matters more than hardware context. To handle data larger than GPU memory, we extend TorchPlan with Dask-cuDF for on-demand partition loading on TPC-H SF100 with four H100 GPUs, achieving $2.54\times$ speedup. Project page: https://kerneldf.github.io/datakernelbench

cs.CL↗

arXiv2026

A Community Survey on SHACL and ShEx: Briding Gaps in RDF Validation

This paper examines RDF validation practices and challenges to understand stakeholder applications, their needs, and identify areas for improvement in technologies and methodologies, thereby guiding future research and standardization efforts. A community survey was conducted, targeting a diverse group of RDF validation technology users across academia and industry. The survey collected data on current practices, tool usage, perceived benefits, limitations, and desired enhancements to gain a broad overview of the validation landscape. Our analysis shows that while RDF validation is widely adopted and valued for enhancing data quality, significant challenges remain. In particular, users report a need for better documentation, improved tool support, enhanced performance, and greater language expressiveness to handle complex large-scale validation tasks effectively. This work provides crucial insights into the RDF validation landscape, highlighting current practices and key areas for development. It offers a foundation for researchers, developers, and standardization bodies to address current limitations and advance validation technologies, ultimately improving data quality and usability in knowledge graphs.

cs.DB↗

arXiv2026

Amalgam: Hybrid LLM-PGM Synthesis Algorithm for Accuracy and Realism

To generate synthetic datasets, e.g., in domains such as healthcare, the literature proposes approaches of two main types: Probabilistic Graphical Models (PGMs) and Deep Learning models, such as LLMs. While PGMs produce synthetic data that can be used for advanced analytics, they do not support complex schemas and datasets. LLMs on the other hand, support complex schemas but produce skewed dataset distributions, which are less useful for advanced analytics. In this paper, we therefore present Amalgam, a hybrid LLM-PGM data synthesis algorithm supporting both advanced analytics, realism, and tangible privacy properties. We show that Amalgam synthesizes data with an average 91 % $\chi^2 P$ value and scores 3.8/5 for realism using our proposed metric, where state-of-the-art is 3.3 and real data is 4.7.

cs.DB↗

arXiv2026

Graph-Native Normalization

In recent years, knowledge graphs (KGs) - in particular in the form of labeled property graphs (LPGs) - have become essential components in a broad range of applications. Although the absence of strict schemas for KGs facilitates structural issues that lead to redundancies and subsequently to inconsistencies and anomalies, the problem of KG quality has so far received only little attention. Inspired by normalization using functional dependencies for relational data, a first approach exploiting dependencies within nodes has been proposed. However, real-world KGs also expose functional dependencies involving edges. In this paper, we therefore propose graph-native normalization, which considers dependencies within nodes, edges, and their combination. We define a range of graph-native normal forms and graph object functional dependencies and propose algorithms for transforming graphs accordingly. We evaluate our contributions using a broad range of synthetic and native graph datasets.

cs.DB↗

arXiv2025

Source Attribution in Retrieval-Augmented Generation

While attribution methods, such as Shapley values, are widely used to explain the importance of features or training data in traditional machine learning, their application to Large Language Models (LLMs), particularly within Retrieval-Augmented Generation (RAG) systems, is nascent and challenging. The primary obstacle is the substantial computational cost, where each utility function evaluation involves an expensive LLM call, resulting in direct monetary and time expenses. This paper investigates the feasibility and effectiveness of adapting Shapley-based attribution to identify influential retrieved documents in RAG. We compare Shapley with more computationally tractable approximations and some existing attribution methods for LLM. Our work aims to: (1) systematically apply established attribution principles to the RAG document-level setting; (2) quantify how well SHAP approximations can mirror exact attributions while minimizing costly LLM interactions; and (3) evaluate their practical explainability in identifying critical documents, especially under complex inter-document relationships such as redundancy, complementarity, and synergy. This study seeks to bridge the gap between powerful attribution techniques and the practical constraints of LLM-based RAG systems, offering insights into achieving reliable and affordable RAG explainability.

cs.LG↗

arXiv2025

MultiHal: Multilingual Dataset for Knowledge-Graph Grounded Evaluation of LLM Hallucinations

Large Language Models (LLMs) have inherent limitations of faithfulness and factuality, commonly referred to as hallucinations. Several benchmarks have been developed that provide a test bed for factuality evaluation within the context of English-centric datasets, while relying on supplementary informative context like web links or text passages but ignoring the available structured factual resources. To this end, Knowledge Graphs (KGs) have been identified as a useful aid for hallucination mitigation, as they provide a structured way to represent the facts about entities and their relations with minimal linguistic overhead. We bridge the lack of KG paths and multilinguality for factual language modeling within the existing hallucination evaluation benchmarks and propose a KG-based multilingual, multihop benchmark called MultiHal framed for generative text evaluation. As part of our data collection pipeline, we mined 140k KG-paths from open-domain KGs, from which we pruned noisy KG-paths, curating a high-quality subset of 25.9k. Our baseline evaluation shows an absolute scale improvement by approximately 0.12 to 0.36 points for the semantic similarity score, 0.16 to 0.36 for NLI entailment and 0.29 to 0.42 for hallucination detection in KG-RAG over vanilla QA across multiple languages and multiple models, demonstrating the potential of KG integration. We anticipate MultiHal will foster future research towards several graph-based hallucination mitigation and fact-checking tasks.

cs.CL↗

arXiv2025

The Limits of Graph Samplers for Training Inductive Recommender Systems: Extended results

Inductive Recommender Systems are capable of recommending for new users and with new items thus avoiding the need to retrain after new data reaches the system. However, these methods are still trained on all the data available, requiring multiple days to train a single model, without counting hyperparameter tuning. In this work we focus on graph-based recommender systems, i.e., systems that model the data as a heterogeneous network. In other applications, graph sampling allows to study a subgraph and generalize the findings to the original graph. Thus, we investigate the applicability of sampling techniques for this task. We test on three real world datasets, with three state-of-the-art inductive methods, and using six different sampling methods. We find that its possible to maintain performance using only 50% of the training data with up to 86% percent decrease in training time; however, using less training data leads to far worse performance. Further, we find that when it comes to data for recommendations, graph sampling should also account for the temporal dimension. Therefore, we find that if higher data reduction is needed, new graph based sampling techniques should be studied and new inductive methods should be designed.

cs.IR↗

arXiv2025

Towards Computer-Using Personal Agents

Computer-Using Agents (CUA) enable users to automate increasingly-complex tasks using graphical interfaces such as browsers. As many potential tasks require personal data, we propose Computer-Using Personal Agents (CUPAs) that have access to an external repository of the user's personal data. Compared with CUAs, CUPAs offer users better control of their personal data, the potential to automate more tasks involving personal data, better interoperability with external sources of data, and better capabilities to coordinate with other CUPAs in order to solve collaborative tasks involving the personal data of multiple users.

cs.HC↗

arXiv2024

Semantic Web: Past, Present, and Future (with Machine Learning on Knowledge Graphs and Language Models on Knowledge Graphs)

Ever since the vision was formulated, the Semantic Web has inspired many generations of innovations. Semantic technologies have been used to share vast amounts of information on the Web, enhance them with semantics to give them meaning, and enable inference and reasoning on them. Throughout the years, semantic technologies, and in particular knowledge graphs, have been used in search engines, data integration, enterprise settings, and machine learning. In this paper, we recap the classical concepts and foundations of the Semantic Web as well as modern and recent concepts and applications, building upon these foundations. The classical topics we cover include knowledge representation, creating and validating knowledge on the Web, reasoning and linking, and distributed querying. We enhance this classical view of the so-called ``Semantic Web Layer Cake'' with an update of recent concepts. These include provenance, security and trust, as well as a discussion of practical impacts from industry-led contributions. We also provide an overiew of shallow and deep machine learning methods for knowledge graphs and discuss the relation of language models and knowledge graphs. We conclude with an outlook on the future directions of the Semantic Web.

cs.AI↗

arXiv2024

Knowledge Graphs, Large Language Models, and Hallucinations: An NLP Perspective

Large Language Models (LLMs) have revolutionized Natural Language Processing (NLP) based applications including automated text generation, question answering, chatbots, and others. However, they face a significant challenge: hallucinations, where models produce plausible-sounding but factually incorrect responses. This undermines trust and limits the applicability of LLMs in different domains. Knowledge Graphs (KGs), on the other hand, provide a structured collection of interconnected facts represented as entities (nodes) and their relationships (edges). In recent research, KGs have been leveraged to provide context that can fill gaps in an LLM understanding of certain topics offering a promising approach to mitigate hallucinations in LLMs, enhancing their reliability and accuracy while benefiting from their wide applicability. Nonetheless, it is still a very active area of research with various unresolved open problems. In this paper, we discuss these open challenges covering state-of-the-art datasets and benchmarks as well as methods for knowledge integration and evaluating hallucinations. In our discussion, we consider the current use of KGs in LLM systems and identify future directions within each of these challenges.

cs.CL↗

arXiv2024

The Susceptibility of Example-Based Explainability Methods to Class Outliers

This study explores the impact of class outliers on the effectiveness of example-based explainability methods for black-box machine learning models. We reformulate existing explainability evaluation metrics, such as correctness and relevance, specifically for example-based methods, and introduce a new metric, distinguishability. Using these metrics, we highlight the shortcomings of current example-based explainability methods, including those who attempt to suppress class outliers. We conduct experiments on two datasets, a text classification dataset and an image classification dataset, and evaluate the performance of four state-of-the-art explainability methods. Our findings underscore the need for robust techniques to tackle the challenges posed by class outliers.

cs.LG↗

arXiv2024

RDFGraphGen: An RDF Graph Generator based on SHACL Shapes

Developing and testing modern RDF-based applications often requires access to RDF datasets with certain characteristics. Unfortunately, it is very difficult to publicly find domain-specific knowledge graphs that conform to a particular set of characteristics. Hence, in this paper we propose RDFGraphGen, an open-source RDF graph generator that uses characteristics provided in the form of SHACL (Shapes Constraint Language) shapes to generate synthetic RDF graphs. RDFGraphGen is domain-agnostic, with configurable graph structure, value constraints, and distributions. It also comes with a number of predefined values for popular schema.org classes and properties, for more realistic graphs. Our results show that RDFGraphGen is scalable and can generate small, medium, and large RDF graphs in any domain.

cs.SE↗

arXiv2024

AIDE: Antithetical, Intent-based, and Diverse Example-Based Explanations

For many use-cases, it is often important to explain the prediction of a black-box model by identifying the most influential training data samples. Existing approaches lack customization for user intent and often provide a homogeneous set of explanation samples, failing to reveal the model's reasoning from different angles. In this paper, we propose AIDE, an approach for providing antithetical (i.e., contrastive), intent-based, diverse explanations for opaque and complex models. AIDE distinguishes three types of explainability intents: interpreting a correct, investigating a wrong, and clarifying an ambiguous prediction. For each intent, AIDE selects an appropriate set of influential training samples that support or oppose the prediction either directly or by contrast. To provide a succinct summary, AIDE uses diversity-aware sampling to avoid redundancy and increase coverage of the training data. We demonstrate the effectiveness of AIDE on image and text classification tasks, in three ways: quantitatively, assessing correctness and continuity; qualitatively, comparing anecdotal evidence from AIDE and other example-based approaches; and via a user study, evaluating multiple aspects of AIDE. The results show that AIDE addresses the limitations of existing methods and exhibits desirable traits for an explainability method.

cs.LG↗

arXiv2023

Hospitalization Length of Stay Prediction using Patient Event Sequences

Predicting patients hospital length of stay (LOS) is essential for improving resource allocation and supporting decision-making in healthcare organizations. This paper proposes a novel approach for predicting LOS by modeling patient information as sequences of events. Specifically, we present a transformer-based model, termed Medic-BERT (M-BERT), for LOS prediction using the unique features describing patients medical event sequences. We performed empirical experiments on a cohort of more than 45k emergency care patients from a large Danish hospital. Experimental results show that M-BERT can achieve high accuracy on a variety of LOS problems and outperforms traditional nonsequence-based machine learning approaches.

cs.LG↗

arXiv2023

KGLiDS: A Platform for Semantic Abstraction, Linking, and Automation of Data Science

In recent years, we have witnessed the growing interest from academia and industry in applying data science technologies to analyze large amounts of data. In this process, a myriad of artifacts (datasets, pipeline scripts, etc.) are created. However, there has been no systematic attempt to holistically collect and exploit all the knowledge and experiences that are implicitly contained in those artifacts. Instead, data scientists recover information and expertise from colleagues or learn via trial and error. Hence, this paper presents a scalable platform, KGLiDS, that employs machine learning and knowledge graph technologies to abstract and capture the semantics of data science artifacts and their connections. Based on this information, KGLiDS enables various downstream applications, such as data discovery and pipeline automation. Our comprehensive evaluation covers use cases in data discovery, data cleaning, transformation, and AutoML. It shows that KGLiDS is significantly faster with a lower memory footprint than the state-of-the-art systems while achieving comparable or better accuracy.

cs.LG↗

arXiv2022

Transforming RDF-star to Property Graphs: A Preliminary Analysis of Transformation Approaches -- extended version

RDF and property graph models have many similarities, such as using basic graph concepts like nodes and edges. However, such models differ in their modeling approach, expressivity, serialization, and the nature of applications. RDF is the de-facto standard model for knowledge graphs on the Semantic Web and supported by a rich ecosystem for inference and processing. The property graph model, in contrast, provides advantages in scalable graph analytical tasks, such as graph matching, path analysis, and graph traversal. RDF-star extends RDF and allows capturing metadata as a first-class citizen. To tap on the advantages of alternative models, the literature proposes different ways of transforming knowledge graphs between property graphs and RDF. However, most of these approaches cannot provide complete transformations for RDF-star graphs. Hence, this paper provides a step towards transforming RDF-star graphs into property graphs. In particular, we identify different cases to evaluate transformation approaches from RDF-star to property graphs. Specifically, we categorize two classes of transformation approaches and analyze them based on the test cases. The obtained insights will form the foundation for building complete transformation approaches in the future.

cs.DB↗

arXiv2022

Simple and Powerful Architecture for Inductive Recommendation Using Knowledge Graph Convolutions

Using graph models with relational information in recommender systems has shown promising results. Yet, most methods are transductive, i.e., they are based on dimensionality reduction architectures. Hence, they require heavy retraining every time new items or users are added. Conversely, inductive methods promise to solve these issues. Nonetheless, all inductive methods rely only on interactions, making recommendations for users with few interactions sub-optimal and even impossible for new items. Therefore, we focus on inductive methods able to also exploit knowledge graphs (KGs). In this work, we propose SimpleRec, a strong baseline that uses a graph neural network and a KG to provide better recommendations than related inductive methods for new users and items. We show that it is unnecessary to create complex model architectures for user representations, but it is enough to allow users to be represented by the few ratings they provide and the indirect connections among them without any user metadata. As a result, we re-evaluate state-of-the-art methods, identify better evaluation protocols, highlight unwarranted conclusions from previous proposals, and showcase a novel, stronger baseline for this task.

cs.IR↗