analiza aiqbench
This research introduces ChemOntoRule, a symbolic framework that constrains large language models with deterministic rules and a task-specific ontology to solve chemistry problems with high accuracy. By separating the reasoning engine from the generative model, the system ensures verifiable logic for school-level applications while addressing the opacity of standard LLMs in scientific domains. The approach demonstrates how integrating structured domain knowledge can mitigate hallucinations without relying solely on probabilistic inference.
A Task-Centric Ontology and Deterministic Domain Rules as a Verifiable Core for AI-Assisted Chemistry Problem Solving
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.
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.
15 pages, 4 figures, 5 tables. Proof-of-concept evaluation of a task-centric chemistry ontology and deterministic rule engine on 300 human-authored and manually validated school-level problems. The proposed LLM translation layer and token-efficiency hypothesis are not evaluated in the current study
Artificial Intelligence (cs.AI)
arXiv:2608.26164 [cs.AI]
(or arXiv:2608.26164v1 [cs.AI] for this version)
Text integral sub licența deschisă a publicației · 285 cuvinte
Subiecte conexe
Game Design Creativity evals
LessWrong · 12 oct. 2026
The Alignment Problem: Computer Science Meets Analytic Philosophy
LessWrong · 18 oct. 2026
Evaluating AI Agents: A production blueprint with Strands and AgentCore
AWS · 23 iul. 2026
Mastering Agentic Techniques: AI Agent Customization
NVIDIA · 20 mai 2026