Dr. techn. Dipl.-Ing. Aleksandar Pavlovic, BSc
Favoritenstraße 226, B.3.35
1100 Wien
+43 1 606 68 77-8405
aleksandar.pavlovic@hcw.ac.at
About
I am a Senior Researcher at Hochschule Campus Wien with an interdisciplinary research profile spanning knowledge representation, neuro‑symbolic AI, machine learning, and database systems. My work focuses on bridging symbolic reasoning and neural models to develop scalable, interpretable, and trustworthy AI systems, with a particular emphasis on graph embeddings, logical reasoning, and responsible language technologies.
I completed my doctoral degree in Computer Science at TU Wien, where my research was conducted at the interface of logic‑based reasoning and geometric machine learning. My PhD was awarded sub auspiciis Praesidentis Rei Publicae, the highest academic distinction in Austria, recognizing continuous academic excellence. My work has been published at leading venues in artificial intelligence, machine learning, and data management, including ICLR, KR, NAACL, and VLDB. Notably, my paper ExpressivE: A Spatio‑Functional Embedding for Knowledge Graph Completion received a Spotlight Award at ICLR 2023, placing it among the top 25% of accepted papers and my paper BoxLitE: A Faithful Knowledge Base Embedding Based on Convex Optimization received a Best Poster Award at DL 2026.
From a research perspective, I am particularly interested in:
- Faithful and differentiable reasoning over structured knowledge
- Neuro‑symbolic representations that respect logical constraints
- Geometric and interpretable embedding models
- Scalable AI systems that translate theoretical guarantees into practice
- Responsible and inclusive language technologies, including bias‑aware evaluation and culturally sensitive AI design
Beyond core theoretical contributions, my research consistently aims to produce practically viable systems, combining formal guarantees with efficiency, interpretability, and reproducibility. Check out my GitHub page for open‑source implementations and AI model architectures of published papers.
Awards
| 2026 (July) | Best Poster Award: I received a Best Poster Award for my paper “BoxLitE: A Faithful Knowledge Base Embedding Based On Convex Optimization (Extended Abstract)” presented at the 39th International Workshop on Description Logics (DL 2026). |
| 2026 (March) | Promotio Sub Auspiciis Praesidentis Rei Publicae: I was awarded the Ring of Honor by the President of Austria upon completing my doctoral degree sub auspiciis, the highest academic distinction in the country, reserved for students with an unbroken record of outstanding academic excellence. |
| 2023 | Spotlight Award: I received a Spotlight Award for my paper “ExpressiveE: A Spatio-Functional Embedding For Knowledge Graph Completion” presented at the 11th International Conference on Learning Representations (ICLR 2023). Spotlight papers represent the top 25% of submissions selected for exceptional quality and impact. |
| 2020 | Epilogue Nomination: My master’s thesis was nominated as one of the four best master’s theses at TU Wien for the graduation year. |
| 2019 | Bachelor with Honors: I was awarded the Bachelor with Honors distinction. This distinction is for graduates of an excellence program designed for the top 5% of TU Wien’s students, extending the bachelor’s curriculum by one year with advanced coursework and international research training. |
Selected Invited Talks and Conference Presentations
| 2026 (August) | Integreat Seminar Series (Invited Talk): I gave an invited talk titled „From Logical Inference to Reasoning in Vector Spaces“ at the Integreat Seminar hosted by the University of Oslo, Norway. |
| 2026 (July) | KR 2026: I presented the paper „BoxLitE: A Faithful Knowledge Base Embedding Based on Convex Optimization“ at the 23rd International Conference on Principles of Knowledge Representation and Reasoning in Lisbon, Portugal. |
| 2026 (January) | Institute of Statistical Mathematics (Invited Talk): I gave an invited talk titled „Reshaping Graph Learning: Expressivity, Inductive Reasoning, and Optimization“ at the Institute of Statistical Mathematics (ISM) in Tokyo, Japan. |
| 2025 (November) | KR 2025: I presented the paper „Faithful Differentiable Reasoning with Reshuffled Region-based Embeddings“ at the 22nd International Conference on Principles of Knowledge Representation and Reasoning in Melbourne, Australia. |
| 2024 (April) | NAACL 2024: I presented the paper „SpeedE: Euclidean Geometric Knowledge Graph Embedding Strikes Back“ at the Nations of the Americas Chapter of the Association for Computational Linguistics in Ciudad de Mexico, Mexico. |
| 2023 (April) | VLDB 2023: I presented the paper „SparqLog: A System for Efficient Evaluation of SPARQL 1.1 Queries via Datalog“ at the 49th International Conference on Very Large Data Bases in Guangzhou, China. |
| 2023 (April) | ICLR 2023 (Spotlight Presentation): I gave a spotlight presentation for the paper „ExpressivE: A Spatio-Functional Embedding for Knowledge Graph Completion“ at the 11th International Conference on Learning Representations in Kigali, Rwanda. |
| 2022 (July) | SoftKR Seminar Series (Invited Talk): I was invited to present my research on learning logical rules in vector spaces at the SoftKR Seminar hosted by the Vrije Universiteit Amsterdam, Netherlands. |
Publications
- BoxLitE: A Faithful Knowledge Base Embedding Based on Convex Optimization
Lourenço, B. F., Morgan, H., Ozaki, A., Pavlović, A., Sallinger, E. (2026). BoxLitE: A Faithful Knowledge Base Embedding Based on Convex Optimization. Proceedings of the 23rd International Conference on Principles of Knowledge Representation and Reasoning, 1006-1017.
https://doi.org/10.24963/kr.2026/94 - BoxLitE: A Faithful Knowledge Base Embedding Based on Convex Optimization (Extended Abstract)
Lourenço, B. F., Morgan, H., Ozaki, A., Pavlović, A., Sallinger, E. (2026). BoxLitE: A Faithful Knowledge Base Embedding Based on Convex Optimization (Extended Abstract). Proceedings of the 39th International Workshop on Description Logics.
https://ceur-ws.org/Vol-4230/dl-paper43.pdf - ExpressiveReshufflE: Exploring Reshuffled Region-based Embeddings in Link Prediction
Schimandl, R., Pavlović, A., Schockaert, S. (2026). ExpressiveReshufflE: Exploring Reshuffled Region-based Embeddings in Link Prediction. Proceedings of the 19th Forschungsforum der österreichischen Fachhochschulen. Accepted for publication. - Convexity for Trustworthy Knowledge Graph Embeddings: From Interpretability to Reliability
Morgan, H., Pavlović, A., Ozaki, A., Lourenço, B. F., Sallinger, E. (2026). Convexity for Trustworthy Knowledge Graph Embeddings: From Interpretability to Reliability. Proceedings of the 19th Forschungsforum der österreichischen Fachhochschulen.
https://ffhoarep.fh-ooe.at/handle/123456789/1832 - Faithful Differentiable Reasoning with Reshuffled Region-based Embeddings
Pavlović, A., Sallinger, E. & Schockaert S. (2025). Faithful Differentiable Reasoning with Reshuffled Region-based Embeddings. In M. Ortiz, R. Wassermann, T. Schaub (Eds.), Proceedings of the 22nd International Conference on Principles of Knowledge Representation and Reasoning, 489-499.
https://doi.org/10.24963/kr.2025/48 - SpeedE: Euclidean Geometric Knowledge Graph Embedding Strikes Back
Pavlović, A., & Sallinger, E. (2024). SpeedE: Euclidean Geometric Knowledge Graph Embedding Strikes Back. In K. Duh, H. Gomez, & S. Bethard (Eds.), Findings of the Association for Computational Linguistics: NAACL 2024 (pp. 69–92). Association for Computational Linguistics.
https://doi.org/10.18653/v1/2024.findings-naacl.6 - Raising the Efficiency of Knowledge Graph Embeddings While Respecting Logical Rules
Pavlović, A. & Sallinger, E. (2024). Raising the Efficiency of Knowledge Graph Embeddings While Respecting Logical Rules. Proceedings of the 16th Alberto Mendelzon International Workshop on Foundations of Data Management.
https://ceur-ws.org/Vol-3954/paper3573.pdf - Expressive and Geometrically Interpretable Knowledge Graph Embedding
Pavlović, A. & Sallinger, E. (2024). Expressive and Geometrically Interpretable Knowledge Graph Embedding. Proceedings of the Austrian Symposium on AI, Robotics, and Vision.
https://semsys.ai.wu.ac.at/sites/KG-NeSy/papers/P60.pdf - SparqLog: A System for Efficient Evaluation of SPARQL 1.1 Queries via Datalog
Angles, R., Gottlob, G., Pavlović, A., Pichler, R., & Sallinger, E. (2023). SparqLog: A System for Efficient Evaluation of SPARQL 1.1 Queries via Datalog. Proceedings of the forty-ninth VLDB Endowment, 16(13), 4240–4253.
https://doi.org/10.14778/3625054.3625061 - Building Bridges: Knowledge Graph Embeddings Respecting Logical Rules
Pavlović, A., & Sallinger, E. (2023). Building Bridges: Knowledge Graph Embeddings Respecting Logical Rules. In B. Kimelfeld, M. V. Martinez, & R. Angles (Eds.), Proceedings of the fifteenth Alberto Mendelzon International Workshop on Foundations of Data Management (AMW 2023).
https://ceur-ws.org/Vol-3409/paper9.pdf - ExpressivE: A Spatio-Functional Embedding For Knowledge Graph Completion
Pavlović, A., & Sallinger, E. (2023). ExpressivE: A Spatio-Functional Embedding For Knowledge Graph Completion. In The Eleventh International Conference on Learning Representations (ICLR 2023) (pp. 1–45). OpenReview.net.
https://openreview.net/forum?id=xkev3_np08z - Vadalog: Overview, Extensions and Business Applications
Baldazzi, T., Bellomarini, L., Gerschberger, M., Jami, A., Magnanimi, D., Nissl, M., Pavlović, A. & Sallinger, E. (2022). Vadalog: Overview, Extensions and Business Applications. In Reasoning Web, 161–198.
https://doi.org/10.1007/978-3-031-31414-8_5
Supervisions
- Beyond Standard Language: Do LLMs Disadvantage Dialect Speakers?
Stückler, M. (2026). Beyond Standard Language: Do LLMs Disadvantage Dialect Speakers? [Master’s Thesis, University of Applied Sciences Wiener Neustadt]. WorldCat. - From Translations to Boxes: Convexifying Knowledge Graph Embedding Approaches, TransE and BoxE
Morgan, H. H. A. A. (2026). From Translations to Boxes: Convexifying Knowledge Graph Embedding Approaches, TransE and BoxE [Diploma Thesis, Technische Universität Wien]. reposiTUm.
https://doi.org/10.34726/hss.2026.139285 - Mining Rules on Knowledge Graph Embeddings: Reducing Rule Set Sizes and Raising Explainability
Gail, D. (2024). Mining Rules on Knowledge Graph Embeddings: Reducing Rule Set Sizes and Raising Explainability [Diploma Thesis, Technische Universität Wien]. reposiTUm.
https://doi.org/10.34726/hss.2024.126768 - Rule Mining on Knowledge Graph Embeddings: Making Implicit Knowledge Explicit and Explainable
Vecera, J. (2024). Rule Mining on Knowledge Graph Embeddings: Making Implicit Knowledge Explicit and Explainable [Diploma Thesis, Technische Universität Wien]. reposiTUm.
https://doi.org/10.34726/hss.2024.119681 - Reasoning in Financial Knowledge Graphs: Making Industry Sectors Accessible to AI
Schüller, M. (2022). Reasoning in Financial Knowledge Graphs: Making Industry Sectors Accessible to AI [Diploma Thesis, Technische Universität Wien]. reposiTUm.
https://doi.org/10.34726/hss.2022.98942 - Recursive Rule Injection in Knowledge Graphs: Exploiting Logical Knowledge in Machine Learning
Wagner, F. K. M. (2021). Recursive Rule Injection in Knowledge Graphs: Exploiting Logical Knowledge in Machine Learning [Diploma Thesis, Technische Universität Wien]. reposiTUm.
https://doi.org/10.34726/hss.2021.85260