Krzysztof Olejniczak
Interests
My research focuses on (automated) neural theorem proving, specifically the autonomous generation of formal proofs in systems such as Lean, and enhancing the formal mathematical reasoning capabilities of AI models. More broadly, I am interested in neuro-symbolic approaches to reasoning and alternative representations for mathematical proof structures.
I also have a background in GNNs and graph representation learning, particularly in the context of reasoning over knowledge graphs.
Selected Publications
-
What are the Right Symmetries for Formal Theorem Proving?
Krzysztof Olejniczak‚ Radoslav Dimitrov‚ Xingyue Huang‚ Bernardo Cuenca Grau‚ Jinwoo Kim and Ismail Ilkan Ceylan
2026.
Details about What are the Right Symmetries for Formal Theorem Proving? | BibTeX data for What are the Right Symmetries for Formal Theorem Proving? | Link to What are the Right Symmetries for Formal Theorem Proving?
-
One Model‚ Any Conjunctive Query: Graph Neural Networks for Answering Queries over Incomplete Knowledge Graphs
Krzysztof Olejniczak‚ Xingyue Huang‚ Mikhail Galkin and İsmail İlkan Ceylan
In Proceedings of Learning on Graphs Conference (LoG). 2025.
Details about One Model‚ Any Conjunctive Query: Graph Neural Networks for Answering Queries over Incomplete Knowledge Graphs | BibTeX data for One Model‚ Any Conjunctive Query: Graph Neural Networks for Answering Queries over Incomplete Knowledge Graphs