A Task-Centric Ontology and Deterministic Domain Rules as a Verifiable Core for AI-Assisted Chemistry Problem Solving

Researchers introduced ChemOntoRule, a proof-of-concept symbolic core for AI-assisted chemistry problem solving that combines a task-centric ontology with deterministic Python rules, and reported that it matched 296 of 300 reference answers (98.67%) on a test set of human-authored chemistry problems. The ontology-driven rule subset covered 269 problems with 266 matches (98.88%), while 31 problems were handled by expert-coded fallbacks with 30 matches. The authors caution that results measure implemented coverage and internal consistency, not independent generalization, and propose a future architecture where a language model acts primarily as a translator into a normalized ontological task frame.

arXiv:2608.26164v1 Announce Type: new Abstract: Large language models can interpret natural-language chemistry questions, but their internal reasoning is difficult to inspect, constrain, and validate. This paper presents ChemOntoRule, a proof-of-concept symbolic core for AI-assisted school-level chemistry problem solving. The central design choice is task-centric ontology engineering: the ontology is constructed around the concepts, properties, relations, and executable procedures required by a defined collection of chemistry problems, rather than as a universal representation of chemistry. The implemented artifact combines a lightweight ontology serialized in JSON and RDF/Turtle with deterministic Python rules for electronic structure, periodic trends, oxidation states, oxide and hydride behavior, and related school-level reasoning patterns. A separate expert-coded fallback handles problem families not yet represented by general rules. The system was examined on 300 human-authored and manually validated chemistry problems. The complete system matched 296 of 300 reference answers 98.67% . The ontology-driven rule subset covered 269 problems and matched 266 references 98.88% ; 31 problems were handled by task-specific expert-coded fallbacks, with 30 matches. Because the same collection informed ontology construction and evaluation, these results measure implemented coverage and internal consistency, not independent generalization. We analyze the four mismatches, distinguish structural validation from chemical correctness, and define a future architecture in which a language model acts primarily as a translator from user language into a normalized ontological task frame. Token efficiency is presented as a testable hypothesis for future controlled studies, not as a result of the current work.