🚀 𝗡𝗲𝘄 𝗥𝗲𝗹𝗲𝗮𝘀𝗲𝘀: 𝗢𝗻𝘁𝗼𝗹𝗲𝗮𝗿𝗻, 𝗢𝗪𝗟𝗔𝗣𝗬 & 𝗗𝗜𝗖𝗘 𝗘𝗺𝗯𝗲𝗱𝗱𝗶𝗻𝗴𝘀

Ontolearn: Refactored concept & neural learners, improved triple store, and new SAT-based learners.👉https://github.com/dice-group/Ontolearn

OWLAPy: Introducing AGen-KG, a scalable agentic KG generation model from text with prompt-based customization.
👉 https://github.com/dice-group/owlapy

DICE Embeddings: New KGE models, ensemble & weight-averaging strategies, improved evaluation workflows, tutorials, and bug fixes.
👉https://github.com/dice-group/dice-embeddings?tab=readme-ov-file

Yesterday, we hosted the HNI Forum 2026 at the Heinz Nixdorf Institute, bringing together researchers to discuss Neurosymbolic Artificial Intelligence.

🎤Prof. Hector Geffner (RWTH Aachen University) spoke on "Representation learning to act and plan”
🎤Prof. Axel-Cyrille Ngonga Ngomo presented “Explainable neurosymbolic machine learning on knowledge graphs”

Great discussions and interdisciplinary exchange—thanks to everyone who joined! 🙌
#NeurosymbolicAI #ExplainableAI #AIResearch #HNI

🚀 We’re excited to announce two paper acceptances The Web Conference (#TheWebConf2026), taking place April 13–17 in Dubai.

👉"A Simplex Approach to Synthetic Knowledge Graph Generation" by Ana Silva, Atul Bhopalsing Pundir, Michael Röder & Axel Ngonga.

👉"Discrete Diffusion-Based Model-Level Explanation of Heterogeneous GNNs with Node Features" by Pallabee Das & Stefan Heindorf.

👏 Congratulations to all authors!

🎙️ New podcast episode (in German): Kernel und Kernobst #073 “Does AI Always Have to Be Friendly?”
@NgongaAxel was a guest and discussed everyday questions about AI, taking a deeper, more academic look at how AI works.
👉https://trommelspeicher.de/podcast/muss-eine-ki-immer-freundlich-sein
#AI #Podcast

We’re excited that our colleague Yasir Mahmood is presenting two papers at the AAAI Conference in Singapore!🚀

👉 "Structure-Aware Encodings of Argumentation Properties for Clique-width" by Yasir Mahmood, Markus Hecher, Johanna Groven & Johannes K. Fichte
👉 "Can You Tell the Difference? Contrastive Explanations for ABox Entailments" by Patrick Koopmann, Yasir Mahmood, Axel Ngona & Balram Tiwari

Wishing you a great time and many inspiring exchanges with the AI community!

#DICEontour #AAAI2026

✨ 𝗛𝗮𝗽𝗽𝘆 𝗛𝗼𝗹𝗶𝗱𝗮𝘆𝘀 𝗳𝗿𝗼𝗺 𝗗𝗜𝗖𝗘✨

As the year comes to an end, we’d like to say thank you to everyone who was part of DICE in 2025, our team, collaborators, partners, and the wider research community.
We wish you happy holidays, a restful end of the year, and a great start to 2026. 🎄✨

✨ 𝗗𝗜𝗖𝗘 𝗶𝗻 𝟮𝟬𝟮𝟱 – 𝗬𝗲𝗮𝗿 𝗶𝗻 𝗥𝗲𝘃𝗶𝗲𝘄 (𝗣𝗮𝗿𝘁 𝟮)

The second half of 2025 brought strong visibility and key milestones for DICE:
🚀 Launch of Learn2RAG (BMWE-funded)
🎉 Public engagement at 100 years Heinz Nixdorf
🌍 Research presented at IJCAI, ECML PKDD, ISWC, K-CAP & more
📈 Our neuro-symbolic ML tools (DICE Embeddings, Ontolearn, OWLAPY) surpassed 100K downloads

Thanks to everyone who made this year a success — here’s to 2026! 🚀

✨ 𝗗𝗜𝗖𝗘 𝗶𝗻 𝟮𝟬𝟮𝟱 – 𝗬𝗲𝗮𝗿 𝗶𝗻 𝗥𝗲𝘃𝗶𝗲𝘄 (𝗣𝗮𝗿𝘁 𝟭)

🌍 We presented DICE research at many international conferences, starting with COLING in Abu Dhabi and including ESWC in Portorož.
🤝 We welcomed Prof. Tibebe Beshah Tesema as a SAIL Fellow in Paderborn.
👥 Our group continued to grow with new members.
🚀 The EU project ENEXA successfully concluded.
🎓 A major milestone was the successful PhD defence of N’Dah Jean Kouagou.

More highlights coming soon in Part 2!

"OWLAPY: A Pythonic Framework for OWL Ontology Engineering" by Alkid Baci, Luke Friedrichs, Caglar Demir & Axel Ngonga is now online! 👉Check it out: https://arxiv.org/abs/2511.08232

🔎 OWLAPY is a comprehensive Python framework for OWL ontology engineering and is already surpassing 50k downloads!

🚀 You can find OWLAPY here: https://github.com/dice-group/owlapy

🎉𝗧𝘄𝗼 𝗽𝗮𝗽𝗲𝗿𝘀 𝗮𝗰𝗰𝗲𝗽𝘁𝗲𝗱 𝗮𝘁 𝗔𝗔𝗔𝗜-𝟮𝟬𝟮𝟲🎉
📄"Can You Tell the Difference? Contrastive Explanations for ABox Entailments" by Patrick Koopmann, Yasir Mahmodd, Axel Ngonga & Balram Tiwari.

📄"Structure-Aware Encodings of Argumentation Properties for Clique-width" by Yasir Mahmood, Markus Hecher, Johanna Groven, and Johannes K. Fichte.

👏 Congratulations to all authors! We are excited to present and discuss our research at one of the world’s leading AI conferences.

#AAAI2026

This week, our colleagues are presenting their latest research @iswc_conf in Nara, Japan!😊 We’re excited to share and discuss our research and to connect with the global Semantic Web community!🤝

#ISWC2025 #SemanticWeb #DICEontour

🎉 Milestone alert: 100K+ downloads!
Our neuro-symbolic ML suite has just surpassed 100,000 downloads! 🚀
These open-source tools power scalable representation learning, concept learning, and reasoning on knowledge graphs.
👉DICE Embeddings - large-scale KG embeddings: https://github.com/dice-group/dice-embeddings
👉Ontolearn – OWL class-expression learning: https://github.com/dice-group/Ontolearn
👉OWLAPY – ontology processing & reasoning: https://github.com/dice-group/owlapy

Thanks to everyone contributing & using these tools 💪

Two more papers accepted for #KCAP2025 🥳

1️⃣ “NL2LS: LLM-based Automatic Linking of Knowledge Graphs” by Reda Ihtassine, Asep Fajar Firmansyah, Nikit Srivastava, Ali Manzoor, Axel Ngonga & Mohamed Ahmed Sherif.

2️⃣ “Neural Reasoning for Robust Instance Retrieval in SHOIQ” by Louis Mozart Kamden, Luke Friedrichs, N'Dah Kouagou, Caglar Demir, Yasir Mahmood, Stefan Heindorf & Axel Ngonga.

We now have a total of 5 DICE papers that will be presented at K-CAP in Dayton/Ohio in December.🙌

New DICE paper! 🎉

"Resilience in Knowledge Graph Embeddings" by Arnab Sharma, N'Dah Jean Kouagou & Axel-Cyrille Ngonga Ngomo

👉We propose a unified definition of resilience and survey current work on robust & adaptive KGE models.

🔎Full paper: https://drops.dagstuhl.de/entities/document/10.4230/TGDK.3.2.1

🚀 Tentris v1.5.0 released! Our research prototype Tentris, a worst-case optimal join triple store, has just received a major update. The new version now supports fast INSERT/DELETE DATA operations — making worst-case optimal joins practically usable in real-world knowledge graph workflows.🤩

👉Check it out: https://github.com/dice-group/tentris-research-project/releases/tag/v1.5.0

#KnowledgeGraphs #SemanticWeb

🎉 3 DICE papers accepted at #KCap2025 (Dec 10–12, Dayton, Ohio)!
Congrats to all authors 👏

1️⃣ "Explainable Benchmarking through the Lens of Concept Learning" by Quannian Zhang, Michael Röder, Nikit Srivastava, N'Dah Jean Kouagou & Axel Ngonga.
2️⃣ "Parameter Averaging in Link Prediction" by Rupesh Sapkota, Caglar Demir, Arnab Sharma & Axel Ngonga.
3️⃣ "Evaluation of Entity and Relation Linking for Question Answering over Knowledge Graphs" by Daniel Vollmers, René Speck, Hamada Zahera & Axel Ngonga.

Our colleague Moshood Yekini presented the paper "Tree-based OWL Class Expression Learner over Large Knowledge Graphs" by C. Demir, M. Yekini, M. Röder, Y. Mahmood & A. Ngonga at #ECMLPKDD2025 in Porto!🇵🇹 👏
👉Preprint: https://ecmlpkdd-storage.s3.eu-central-1.amazonaws.com/preprints/2025/research/preprint_ecml_pkdd_2025_research_470.pdf

#DICEontour #KnowledgeGraphs

New paper!🙌 "Box embeddings for extending ontologies: a data-driven and interpretable approach”, by Adel Memariani, Martin Glauer, Simon Flügel, Fabian Neuhaus, Janna Hastings & Till Mossakowski. ➡️Read the full paper published in Journal of Cheminformatics here: https://doi.org/10.1186/s13321-025-01086-1

We were thrilled to join the Lamarr Scientific Forum!🤩 Duygu Ekinci & Shivam Sharma presented the poster “Towards Universal Knowledge Graph Embeddings”, showcasing scalable & explainable AI for knowledge graphs. Thank you to the organizers of
the Lamarr Institute! 👏 #DICEontour

We are delighted that two papers will be presented at the Wikidata Workshop @iswc_conf !🥳

1) “MST5 - A Transformers-Based Approach to Multilingual Question Answering over Knowledge Graphs” by Nikit Srivastava, Mengshi Ma, Daniel Vollmers, Hamada Zahera, Diego Moussallem & Axel Ngonga

2) “LoSA: A Local Structural Approach to Adversarial Attack on the Knowledge Graph-based Question Answering System” by Neha Pokarel, Michael Röder, Adel Memariani, Arnab Sharma & Axel Ngonga

Congratulations!👏