Days: Tuesday, August 11th Wednesday, August 12th Thursday, August 13th Friday, August 14th
View this program: with abstractssession overviewtalk overview
ARDUOUS: 10th International Workshop on Annotation of Real World Data for Artificial Intelligence Systems
https://arduous.eu/
Workshop on AIDriven Cyber Forensics (AICF)
https://ki2026.polizeiinformatik.de/
Doctoral Consortium
https://ki2026.gi.de/calls/call-for-doctoral-consortium-contributions
| 09:00 | Toward Task-Sensitive Human-in-the-Loop AI-DSS: Configuring System Mechanisms for Advice Acceptance in Supply Chain and Manufacturing (abstract) |
| 09:30 | Trust as a Mediator of AI Acceptance in Insurance Advice (abstract) |
| 10:00 | Learnable Fairness and Explainability in Social Combinatorial Optimization (abstract) |
ARDUOUS: 10th International Workshop on Annotation of Real World Data for Artificial Intelligence Systems
https://arduous.eu/
Workshop on AIDriven Cyber Forensics (AICF)
https://ki2026.polizeiinformatik.de/
Doctoral Consortium
https://ki2026.gi.de/calls/call-for-doctoral-consortium-contributions
| 11:00 | Tutorial: On writing research papers (abstract) |
ARDUOUS: 10th International Workshop on Annotation of Real World Data for Artificial Intelligence Systems
https://arduous.eu/
ExTraSafe - 1st KI Workshop on Explainability, Transparency, and Safety
https://fb-ki.gi.de/extrasafe
Doctoral Consortium
https://ki2026.gi.de/calls/call-for-doctoral-consortium-contributions
| 14:00 | Epistemically Grounded Neuro-Symbolic Coordination in Large Language Model-Based Multi-Agent Systems (abstract) |
| 14:30 | Reference-Free Automated Prompt Optimization for RAG-Based Microlearning Generation (abstract) |
| 15:00 | Automatic Extraction of Constraints from Machine Learning Models for Duplication Detection (abstract) |
ARDUOUS: 10th International Workshop on Annotation of Real World Data for Artificial Intelligence Systems
https://arduous.eu/
ExTraSafe - 1st KI Workshop on Explainability, Transparency, and Safety
https://fb-ki.gi.de/extrasafe
Doctoral Consortium
https://ki2026.gi.de/calls/call-for-doctoral-consortium-contributions
| 16:00 | Understanding Multimodal Interference through Local Preference Inconsistency (abstract) |
| 16:30 | Mitigating Data Scarcity in Lifecycle Data using AI and Matrix Completion Techniques: A Recommender System Design (abstract) |
| 17:00 | Hypergraph Provenance Causal Learning for APT Detection in the Internet of Medical Things (abstract) PRESENTER: Uthayanathan Priyatharsan |
View this program: with abstractssession overviewtalk overview
Workshop Applications of Artificial Intelligence for Software Engineering (AI4SE)
https://siebert-julien.github.io/ai4se-workshop/
See: https://gesina.github.io/ki2026-c-xai-tutorial/
Workshop SIG Knowledge Management
https://fg-wm.wi2.uni-trier.de/
Workshop Applications of Artificial Intelligence for Software Engineering (AI4SE)
https://siebert-julien.github.io/ai4se-workshop/
Workshop SIG Knowledge Management
https://fg-wm.wi2.uni-trier.de/
Conference opening and welcome address
Keynote
| 14:00 | Re-emerging scientific challenges of Artificial Intelligence |
Poster spotlight #1
| 15:00 | A Hybrid Approach for Generating Planning Models from Texts (abstract) PRESENTER: Teodor Stoev |
| 15:05 | PictSure: Pretraining Embeddings Matters for In-Context Learning Image Classifiers (abstract) |
| 15:10 | A Comparison of Repositioning and Scheduling Algorithms for the Ride-Hailing Problem (abstract) |
| 15:15 | Comparison of the Runtime of two Algorithms for the Linear Decomposition of ReLU Networks (abstract) |
| 15:20 | Do Flat Representation Manifolds lead to improved Accuracy? (abstract) |
| 15:25 | From Large Language Model Predicates to Logic Tensor Networks: Neurosymbolic Offer Validation in Regulated Procurement (abstract) |
session with paper presentations of full papers
| 16:00 | Achieving Fairness in Repeated Combinatorial Problems through Deep Reinforcement Learning (abstract) |
poster spotlight #2 for technical communications
| 16:30 | Improving WSSED Transfer to PAM via Multi-Species Augmentation (abstract) |
| 16:35 | AI and E-Learning: On the importance of explainable AI in Digital Ecosystems in Research and Education (abstract) |
General assemby of the German AI society (“Fachbereich KI”, FBKI) within the German Informatics Society (GI e.V.) – for members of FBKI but open to guests.
We will take a guided tour through the historic town of Bremen, which will bring us to the conference banquet.
| 20:50 | Innovation and Democracy in the Age of AI |
View this program: with abstractssession overviewtalk overview
| 09:00 | Can We Trust Generative AI? Understanding and Mitigating Threats in Today’s Machine Learning Systems (abstract) |
Poster spotlight #3 of technical communications
| 10:00 | Can We Predict LLM Reasoning Failures? Structural Predictability of Modal Inference Errors (abstract) PRESENTER: Fatemeh Shahrokhshahi |
| 10:05 | Quantum Temporal Convolution Network (abstract) |
| 10:10 | Qualitative Comparison between Marker-Based and Video-Based Human Pose Estimation in the context of Imitation Learning (abstract) |
| 10:15 | An Ontology-Grounded Representation for Defeasible Professional Ethics Analysis (abstract) |
| 10:20 | Safeguards for Speech2Speech LLM-Assistants: A Case Study in Automotive Applications (abstract) |
| 10:25 | WaLo2D: Testbed for Multi-Agent Warehouse Logistics Reinforcement Learning Experiments (abstract) |
full paper presentations
| 11:00 | A Constraint-based Stockyard Planning Problem on a Realistic Time-Dependent Setting (abstract) |
| 11:30 | Prioritizing Conflicts in Conflict-Based Search with Disjoint Splitting (abstract) |
| 12:00 | Integrating Insertion Heuristics into the Miller-Tucker-Zemlin Constraint Model for the Traveling Salesperson Problem (abstract) |
Poster spotlight #4 of technical communications
| 14:00 | Explainable-AI-Based Training for Relevance-based Robust Reinforcement Learning (abstract) |
| 14:05 | Constrained Entity Selection under Partial Knowledge for LLM-Based Knowledge Graph QA (abstract) |
| 14:10 | Evaluating AI as Part of Social Mechanisms (abstract) |
| 14:15 | Neuro-Symbolic Verification of LLM Outputs for Data-Sensitive Domains (abstract)  PRESENTER: Paul Sigloch |
full paper presentations
| 14:30 | SIL-RRT*: Learning Sampling Distribution through Self Imitation Learning (abstract) |
| 15:00 | Efficient Time-Series Approximation with Linear Recurrent Neural Networks (abstract) |
The industial session aims to connect research with industrial applications of AI. The session will take place in the historic rooms of the Bremen Chamber of Commerce (house “Schütting” at the central town square) and will start with the keynote of Michael Spranger from Sony AI. It will be continue with pitches of AI Use Cases from industry followed with the possibility for bilatral discussions, exchanges and networking.
- Buhlmann Group: AI Adoption at BUHLMANN Group by 1.000+ People
- Contact Software GmbH: Accelerating Engineering with AI: Virtual Assistance, 3D Similarity Search & Impact Analysis
- Fuseki GmbH: Practical AI Implementation in Companies
- HEC Hanseatische Software-Entwicklungs- und Consulting-Gesellschaft mbH: AI-supported Order Processing in Logistics for Oetjen Logistik GmbH
- Lenze SE /Encoway GmbH: GenAI for Complex Industrial Products
- ma-co maritimes competenzcentrum GmbH: AI Transformation in the Port Industry
- Mercedes Benz Plant Bremen: Innovation made in Bremen: How the Mercedes-Benz Plant is re-designing global digital quality control systems
The webpage with more details is available at the website of the Bremen chamber of commerce: Industrial Event detailed program. Participants of the KI2026 conference do not need to register there as it is part of the conference.
| 16:30 | The Role of Artificial Intelligence at Sony |
At the poster session the posters teased in the poster spotlight talks will be presented. It will take place on board of a ship while we cruise along the Weser river. Please transfer directly from the industrial session to the ship as departure is 20:00 sharp(!).
And don't forget: Vote for your favorite poster(s)!
Short paper posters:
- Teodor Stoev and Kristina Yordanova. A Hybrid Approach for Generating Planning Models from Texts
- Lukas Schiesser, Cornelius Wolff, Sophie Haas and Simon Pukrop. PictSure: Pretraining Embeddings Matters for In-Context Learning Image Classifiers
- Lauren Paul, Sebastian Hübner, Christina Plump, Bernhard J. Berger and Rolf Drechsler. A Comparison of Repositioning and Scheduling Algorithms for the Ride-Hailing Problem
- Arn Dietz, Simon Lutz, Martin Brenzke, Thea Radüntz and Emmanuel Müller. Comparison of the Runtime of two Algorithms for the Linear Decomposition of ReLU Networks
- Julia Burr and Alex Sarishvili. Do Flat Representation Manifolds lead to improved Accuracy?
- Cedric Haufe and Frieder Stolzenburg. From Large Language Model Predicates to Logic Tensor Networks: Neurosymbolic Offer Validation in Regulated Procurement
- Novruz Mammadli, Thiago S. Gouvêa and Daniel Sonntag. Improving WSSED Transfer to PAM via Multi-Species Augmentation
- Claudia-Melania Chituc. AI and E-Learning: On the importance of explainable AI in Digital Ecosystems in Research and Education
- Rihab Hoceini and Ahmed Bouida. Quantum Temporal Convolution Network
- Fatemeh Shahrokhshahi and Farzan Mohammadi. Can We Predict LLM Reasoning Failures? Structural Predictability of Modal Inference Errors
- Max Lödige, Lisa Gutzeit and Alexander Fabisch. Qualitative Comparison between Marker-Based and Video-Based Human Pose Estimation in the context of Imitation Learning
- Christopher Rauch. An Ontology-Grounded Representation for Defeasible Professional Ethics Analysis
- Gregor Endler, Sebastian Kraus and Lukas Stappen. Safeguards for Speech2Speech LLM-Assistants: A Case Study in Automotive Applications
- Christoph Saad and Stefan Geißelsöder. WaLo2D: Testbed for Multi-Agent Warehouse Logistics Reinforcement Learning Experiments
- Benedikt Schlereth-Groh, Sakir Furkan Yöndem, Ramin Tavakoli Kolagari and Ute Schmid. Explainable-AI-Based Training for Relevance-based Robust Reinforcement Learning
- Emanuel Kitzelmann. Constrained Entity Selection under Partial Knowledge for LLM-Based Knowledge Graph QA
- Susanne Draheim, Ulrich Furbach and Ralf Möller. Evaluating AI as Part of Social Mechanisms
- Paul Sigloch and Christoph Benzmüller. Neuro-Symbolic Verification of LLM Outputs for Data-Sensitive Domains
Additional posters of full papers:
- Stefan Bühler and Mark Schutera. !Imperio, smolVLA: The Implications of Data Poisoning on Open Source Robotics
- Tobias Tefke and Ralf C. Staudemeyer. Successfully Defending Academic Integrity: an Ensemble Approach Towards Detecting Machine-generated Texts
- Thomas Vitry, Kieran Edgeworth, Stefan Wermter and Jae Hee Lee. Bias Leaves a Gradient Trail: Label-Free Bias Identification via Gradient Probes on Concept Decompositions
- Simin Mirzadeh, Shamim Miroliaei, Alexej Zitzer, Johannes Munk, Matthias Gempel and Ute Schmid. Multi-State PatchCore for Noisy Industrial Audio
Posters of the doctoral consortium:
- Mayar Abdelwahab: Epistemically Grounded Neuro-Symbolic Coordination in Large Language Model-Based Multi-Agent Systems
- Uthayanathan Priyatharsan: Hypergraph Provenance Causal Learning for APT Detection in the Internet of Medical Things
- Xuerui Zhang: Understanding Multimodal Interference through Local Preference Inconsistency
- André Hülsdau: Trust as a Mediator of AI Acceptance in Insurance Advice
- Fabian Schlagheck: Reference-Free Automated Prompt Optimization for RAG-Based Microlearning Generation
View this program: with abstractssession overviewtalk overview
| 09:00 | Geometric Representations of Relational Knowledge for Neuro-Symbolic Reasoning (abstract) |
| 10:00 | Bias Leaves a Gradient Trail: Label-Free Bias Identification via Gradient Probes on Concept Decompositions (abstract) |
| 11:00 | c-Core Revision for Conditional Belief Bases (abstract) PRESENTER: Marco Wilhelm |
| 11:30 | Characteristics, Convergence and Divergence of Modular Semantics for Bipolar Weighted Argumentation Graphs (abstract) |
| 12:00 | !Imperio, smolVLA: The Implications of Data Poisoning on Open Source Robotics (abstract) |
| 14:00 | Multi-State PatchCore for Noisy Industrial Audio (abstract) |
| 14:30 | Do We Really Need Diffusion? A Fast U-Net for Paired Medical Image Translation (abstract) |
| 15:00 | Successfully Defending Academic Integrity: an Ensemble Approach Towards Detecting Machine-generated Texts (abstract) |
| 15:30 | Code Generation for Open Data Statistics: Case Study on the Genesis Database (abstract) |