This book constitutes the proceedings of the 48th German Conference on Artificial Intelligence (Künstliche Intelligenz), KI 2025, which was held in Potsdam, Germany, during September 1619, 2025.

The 15 full papers, 8 short papers and 5 extended abstracts presented in these proceedings were carefully reviewed and selected from 74 submissions. They focus on new research results on theory and applications in AI. The papers were categorized in the following sections: Full Technical Papers; Technical Communications; Extended Abstracts.

Klappentext

.- Full Technical Papers. .- Augmenting Systematic Literature Reviews: A Human-AI Collaborative Framework. .- Balanced Reciprocity for Data Sharing - Axiomatization and Mechanism Design. .- Toward Short and Robust Contrastive Explanations for Image Classification by Leveraging Instance Similarity and Concept Relevance. .- Intermediate-Task Transfer Learning for Bioacoustic Data. .- On the Domain Robustness of Contrastive Vision-Language Models. .- Numbers Don’t Lie: Hybrid Extraction and Validation of Quantitative Statements in Arguments with Semi-Structured Information. .- A Hybrid Constraint-Based, Greedy, and Local Search Approach for the Transshipment Problem. .- Re-examining learning linear functions in context. .- ODExAI: A Comprehensive Object Detection Explainable AI Evaluation. .- Enhancing Semi-Supervised Learning with a Meta-Feature Based Safeguard System. .- Ca¨ ssa AI: A Neuro-Symbolic Chess Agent for Explainable Move Suggestion and Grounded Commentary. .- Unsupervised Selection of Features by their Resilience to the Curse of Dimensionality. .- Development of Hybrid Artificial Intelligence Training on Real and Synthetic Data. .- Towards Systematic Evaluation of Computer Vision Models under Data Anonymization. .- Accessible Language Simplification: Large Language Models for Generating Easy German. .- Technical Communications. .- Learn, Optimize, Explain: A Neuro-Symbolic Advisor for Personal Finance. .- Deep learning emulators for large-scale, high-resolution urban pluvial flood prediction. .- Towards Observing the Effect of Abstraction on Understandability of Explanations in Answer Set Programming. .- XAIRob — An Explainable-AI-Based Relative Robustness Measure for Object Detection. .- LLMs for Easy Language Translation: A Case Study on German Public Authorities Web Pages. .- Re-Evaluating the Robustness and Interpretability of the Contrastive Explanations Method for Image Classification. .- Visualizing and Interpreting Neural Network Focus Regions: A Comparative Study of Vision Transformers on Synthetic and Real Data. .- Comparing the visual quality of deep generative models for steel microstructures. .- Extended Abstracts. .- Makrut Attacks Against Black-Box Explanations. .- Positional Overload: Positional Debiasing and Context Window Extension for Large Language Models using Set Encoding. .- Exploiting Contexts of LLM-based Code-Completion. .- The origins of AI research in the Federal Republic of Germany. .- Probabilities of the Third Type: Statistical Relational Learning and Reasoning with Relative Frequencies.

Inhalt

.- Full Technical Papers.

.- Augmenting Systematic Literature Reviews: A Human-AI Collaborative
Framework.

.- Balanced Reciprocity for Data Sharing - Axiomatization and
Mechanism Design.

.- Toward Short and Robust Contrastive Explanations for Image
Classification by Leveraging Instance Similarity and Concept Relevance.

.- Intermediate-Task Transfer Learning for Bioacoustic Data.

.- On the Domain Robustness of Contrastive Vision-Language Models.

.- Numbers Don't Lie: Hybrid Extraction and Validation of Quantitative
Statements in Arguments with Semi-Structured Information.

.- A Hybrid Constraint-Based, Greedy, and Local Search Approach for
the Transshipment Problem.

.- Re-examining learning linear functions in context.

.- ODExAI: A Comprehensive Object Detection Explainable AI Evaluation.

.- Enhancing Semi-Supervised Learning with a Meta-Feature Based
Safeguard System.

.- Ca¨ssa AI: A Neuro-Symbolic Chess Agent for Explainable Move
Suggestion and Grounded Commentary.

.- Unsupervised Selection of Features by their Resilience to the Curse of
Dimensionality.

.- Development of Hybrid Artificial Intelligence Training on Real and
Synthetic Data.

.- Towards Systematic Evaluation of Computer Vision Models under Data
Anonymization.

.- Accessible Language Simplification: Large Language Models for
Generating Easy German.

.- Technical Communications.

.- Learn, Optimize, Explain: A Neuro-Symbolic Advisor for Personal Finance.

.- Deep learning emulators for large-scale, high-resolution urban pluvial
flood prediction.

.- Towards Observing the Effect of Abstraction on Understandability of
Explanations in Answer Set Programming.

.- XAIRob An Explainable-AI-Based Relative Robustness Measure for
Object Detection.

.- LLMs for Easy Language Translation: A Case Study on German Public
Authorities Web Pages.

.- Re-Evaluating the Robustness and Interpretability of the Contrastive
Explanations Method for Image Classification.

.- Visualizing and Interpreting Neural Network Focus Regions: A
Comparative Study of Vision Transformers on Synthetic and Real Data.

.- Comparing the visual quality of deep generative models for steel
microstructures.

.- Extended Abstracts.

.- Makrut Attacks Against Black-Box Explanations.

.- Positional Overload: Positional Debiasing and Context Window
Extension for Large Language Models using Set Encoding.

.- Exploiting Contexts of LLM-based Code-Completion.

.- The origins of AI research in the Federal Republic of Germany.

.- Probabilities of the Third Type: Statistical Relational Learning and
Reasoning with Relative Frequencies.