OrLog: Resolving Complex Queries with LLMs and Probabilistic Reasoning
Mohanna Hoveyda, Jelle Piepenbrock, Arjen P. de Vries, Maarten de Rijke, Faegheh Hasibi · 2026
Read OrLog: Resolving Complex Queries with LLMs and Probabilistic Reasoning by Mohanna Hoveyda, Jelle Piepenbrock, Arjen P. de Vries, Maarten de Rijke, Faegheh Hasibi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-23085.txt.
Description
Mohanna Hoveyda and four co-authors answer complex questions that carry strict logical conditions. A query may demand results meeting several constraints joined by and, or, and not. Neural search often ignores those operators, while purely generative reasoning applies them unevenly.
Their system, OrLog, joins language models with probabilistic reasoning to enforce the logic. Candidate answers are checked against the query's structure in a principled way. Reliable constraint handling matters for legal, medical, and product search, where a missed not changes the answer. Probabilistic reasoning supplies the discipline, so answers must satisfy the query's conjunctions, disjunctions, and negations in a consistent way.
Catalog record
| CREATOR | Mohanna Hoveyda, Jelle Piepenbrock, Arjen P. de Vries, Maarten de Rijke, Faegheh Hasibi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 35 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-23085