ENBIS-25 Conference

Session

AI: Interpretability and Trustworthiness

15 Sept 2025, 13:30

Conference Hall (Ground floor)

Conveners

AI: Interpretability and Trustworthiness: 1

• Sonja Kuhnt (Dortmund University of Applied Sciences and Arts)

AI: Interpretability and Trustworthiness: 2

• Lara Kuhlmann de Canaviri (Fachhochschule Dortmund)

44. Selecting Statistical Metrics for Prompt Optimization of Locally Hosted LLMs: A Case Study in Asset Registration

Dr Guido Moeser (masem research institute)

15/09/2025, 13:30

AI: Interpretability and Trustworthiness

In a client project focused on asset registration, we transitioned from a publicly available LLM (Gemini) to a locally hosted LLM to prevent the potential leakage of sensitive manufacturing and customer data. Due to hardware constraints (GPUs with a maximum of 12 GB VRAM), the performance of local LLMs was initially inferior to hosted models. Therefore, prompt optimization became crucial to...

78. Plots for XAI: FANOVA Graph and FaithShapGraph

Lara Kuhlmann de Canaviri (Fachhochschule Dortmund)

15/09/2025, 13:50

AI: Interpretability and Trustworthiness

Explainable AI (XAI) approaches, most notably Shapley values, have become increasingly popular because they reveal how individual features contribute to a model’s predictions. At the same time, global sensitivity analysis (GSA) techniques, especially Sobol indices, have long been used to quantify how uncertainty in each input (and combinations of inputs) propagates to uncertainty in the...

80. Evaluating Personality Consistency in Large Language Models via Big Five Inventories and Dual-Enhanced Networks

Ms Yi-Ting Hou (National Taiwan University)

15/09/2025, 14:10

AI: Interpretability and Trustworthiness

Large language models (LLMs) are increasingly capable of simulating human-like personalities through prompt engineering, presenting novel opportunities and challenges for personality-aware AI applications. However, the consistency and stability of these simulated personalities over time and across contexts remain largely unexplored. In this study, we propose an integrated framework to...

81. Autonomous Persona Generation: Leveraging Agentic AI for Scalable, Adaptive User Simulation

Ms You-Hua Wu (National Taiwan University)

15/09/2025, 14:35

AI: Interpretability and Trustworthiness

This research proposes a novel framework for automating the generation of AI personas using Agentic AI systems, designed to be MCP-compliant for robust interoperability and seamless tool integration. Traditionally, creating AI personas involves repetitive prompt engineering, manual calibration, and continuous human intervention, which can be time-consuming and error-prone. In contrast, we...

85. Weakly Supervised Deep Learning for Disentangling Label and Measurement Noise Using Class-Specific Latent Maps

Dr Bart De Ketelaere (KU Leuven)

15/09/2025, 14:55

AI: Interpretability and Trustworthiness

Deep learning (DL) models are significantly impacted by label and measurement noise, which can degrade performance. Label noise refers to wrong labels (Y) attached to samples, whereas measurement noise refers to the samples (X) that are corrupted due to issues during their acquisition. We present a generic approach for efficient learning in the presence of such noise, without relying on ground...

13. Artificial Intelligence-Enhanced Multivariate Process Control for Gearbox Failure Monitoring via Principal Components Analysis and Machine Learning.

Susana Barceló Cerdá (Universitat Politècnica de València)

15/09/2025, 15:15

Statistical Process Monitoring

This work presents a methodology for condition monitoring of spur gearboxes based on AI-enhanced multivariate statistical process control. Gearboxes are critical components in rotating machinery, and early fault detection is essential to minimize downtime and optimize maintenance strategies. Vibration signals are a non-invasive means to assess gearbox conditions under varying load and...