Joost Vennekens
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Joost Vennekens has a total of 77 co-authors. Here are the top co-authors by number of shared publications:
- Marc Denecker
Department of Computer Science, KU Leuven, Leuven, Belgium
26 shared publications - Simon Vandevelde
KU Leuven
15 shared publications - Bart Bogaerts
KU Leuven
7 shared publications Maurice Bruynooghe
Department of Computer Science, KU Leuven
7 shared publications- Benjamin Callewaert
Dept. of Computer Science KU Leuven De Nayer Campus Belgium
4 shared publications Hanne Vlaeminck
KU Leuven
4 shared publications
Latest Publications
- 23 October 2026
- Published by Springer Nature
- p. 403-422
Can We GLUE It? Extending an Interactive Adhesive Selector with Knowledge Graphs
- 23 October 2026
- Published by Springer Nature
- p. 403-422
Can We GLUE It? Extending an Interactive Adhesive Selector with Knowledge Graphs
- 11 October 2026
- Published by Springer Nature
- p. 448-463
Efficient Grounding with Bounds using Bit Vectors
- 11 October 2026
- Published by Springer Nature
- p. 448-463
Efficient Grounding with Bounds using Bit Vectors
- 12 January 2026
- Published by Springer Nature
- p. 115-132
An Efficient Compiler for the IDP-Z3 Knowledge Base System
- 12 January 2026
- Published by Springer Nature
- p. 115-132
An Efficient Compiler for the IDP-Z3 Knowledge Base System
- Top Cited PapersOpen Access
- 8 January 2026
- Published by Open Publishing Association in Electronic Proceedings in Theoretical Computer Science
- Vol. 439, 47-62
VERUS-LM: a Versatile Framework for Combining LLMs with Symbolic Reasoning
A recent approach to neurosymbolic reasoning is to explicitly combine the strengths of large language models (LLMs) and symbolic solvers to tackle complex reasoning tasks. However, current approaches face significant limitations, including poor generalizability due to task-specific prompts, inefficiencies caused by the lack of separation between knowledge and queries, and restricted inferential capabilities. These shortcomings hinder their scalability and applicability across diverse domains. In this paper, we introduce VERUS-LM, a novel framework designed to address these challenges. VERUS-LM employs a generic prompting mechanism, clearly separates domain knowledge from queries, and supports a wide range of different logical reasoning tasks. This framework enhances adaptability, reduces computational cost, and allows for richer forms of reasoning, such as optimization and constraint satisfaction. We show that our approach succeeds in diverse reasoning on a novel dataset, markedly outperforming LLMs. Additionally, our system achieves competitive results on common reasoning benchmarks when compared to similar state-of-the-art approaches, and significantly surpasses them on the difficult AR-LSAT dataset. By pushing the boundaries of hybrid reasoning, VERUS-LM represents a significant step towards more versatile neurosymbolic AI systems.
- 8 January 2026
- Published by Open Publishing Association in Electronic Proceedings in Theoretical Computer Science
- Vol. 439, 47-62
Top Cited PapersOpen AccessVERUS-LM: a Versatile Framework for Combining LLMs with Symbolic Reasoning
A recent approach to neurosymbolic reasoning is to explicitly combine the strengths of large language models (LLMs) and symbolic solvers to tackle complex reasoning tasks. However, current approaches face significant limitations, including poor generalizability due to task-specific prompts, inefficiencies caused by the lack of separation between knowledge and queries, and restricted inferential capabilities. These shortcomings hinder their scalability and applicability across diverse domains. In this paper, we introduce VERUS-LM, a novel framework designed to address these challenges. VERUS-LM employs a generic prompting mechanism, clearly separates domain knowledge from queries, and supports a wide range of different logical reasoning tasks. This framework enhances adaptability, reduces computational cost, and allows for richer forms of reasoning, such as optimization and constraint satisfaction. We show that our approach succeeds in diverse reasoning on a novel dataset, markedly outperforming LLMs. Additionally, our system achieves competitive results on common reasoning benchmarks when compared to similar state-of-the-art approaches, and significantly surpasses them on the difficult AR-LSAT dataset. By pushing the boundaries of hybrid reasoning, VERUS-LM represents a significant step towards more versatile neurosymbolic AI systems.
- Open Access
- 29 October 2025
- Published by Springer Nature
- p. 258-275
A Domain Ontology for Ishikawa Diagrams to Enhance Root Cause Analysis
- 29 October 2025
- Published by Springer Nature
- p. 258-275
Open AccessA Domain Ontology for Ishikawa Diagrams to Enhance Root Cause Analysis
- 29 August 2025
- Published by Springer Nature
- p. 343-356
DIRT: a Literature-Based Benchmark Suite for Grounders
- 29 August 2025
- Published by Springer Nature
- p. 343-356
DIRT: a Literature-Based Benchmark Suite for Grounders
- 18 March 2025
- Published by Springer Nature
- p. 88-103
A Practical Approach to Handling Tabular Data in Logic
- 18 March 2025
- Published by Springer Nature
- p. 88-103
A Practical Approach to Handling Tabular Data in Logic
- Open Access
Efficiently grounding FOL using bit vectors
Several paradigms for declarative problem solving start from a specification in a high-level language, which is then transformed to a low-level language, such as SAT or SMT. Often, this transformation includes a "grounding" step to remove first-order quantification. To reduce the time and size of the grounding, it can be useful to simplify formulas along the way, e.g., by already taking into account the interpretation of symbols that are already known. In this paper, we investigate the use of bit vectors to efficiently simplify formulas, thereby taking advantage of the fact that, on modern hardware, logical operations on bit vectors can be executed extremely fast. We conduct an experimental analysis, which shows that bit vectors are indeed fast for certain problems, but also have limitations.
Open AccessEfficiently grounding FOL using bit vectors
Several paradigms for declarative problem solving start from a specification in a high-level language, which is then transformed to a low-level language, such as SAT or SMT. Often, this transformation includes a "grounding" step to remove first-order quantification. To reduce the time and size of the grounding, it can be useful to simplify formulas along the way, e.g., by already taking into account the interpretation of symbols that are already known. In this paper, we investigate the use of bit vectors to efficiently simplify formulas, thereby taking advantage of the fact that, on modern hardware, logical operations on bit vectors can be executed extremely fast. We conduct an experimental analysis, which shows that bit vectors are indeed fast for certain problems, but also have limitations.
- Open Access
- 5 June 2024
- Published by IOS Press
- p. 124-133
Towards Effective Management of Verbal Probability Expressions Using a Co-Learning Approach
- 5 June 2024
- Published by IOS Press
- p. 124-133
Open AccessTowards Effective Management of Verbal Probability Expressions Using a Co-Learning Approach