{"paper":{"title":"Ontology-Constrained Neural Reasoning in Enterprise Agentic Systems: A Neurosymbolic Architecture for Domain-Grounded AI Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Ontology-coupled agents significantly outperform ungrounded agents on accuracy and role consistency across enterprise domains.","cross_cats":["cs.CL","cs.SE"],"primary_cat":"cs.AI","authors_text":"Abhijit Sanyal, Thanh Luong Tuan","submitted_at":"2026-04-01T06:59:15Z","abstract_excerpt":"Enterprise adoption of Large Language Models (LLMs) is constrained by hallucination, domain drift, and the inability to enforce regulatory compliance at the reasoning level. We present a neurosymbolic architecture implemented within the Foundation AgenticOS (FAOS) platform that addresses these limitations through ontology-constrained neural reasoning. We introduce a three-layer ontological framework--Role, Domain, and Interaction ontologies--grounding LLM-based enterprise agents. We formalize asymmetric neurosymbolic coupling: current enterprise systems constrain agent inputs (context assembly"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"ontology-coupled agents significantly outperform ungrounded agents on Metric Accuracy (p < .001) and Role Consistency (p < .001) across all three models with large effect sizes (Kendall's W = .46-.64)","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The ontologies are correctly specified and complete for the tested domains, and the controlled experiment adequately isolates the effect of ontological coupling from other factors such as prompt engineering or tool selection.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Ontology grounding improves accuracy and role consistency of enterprise LLM agents, with larger gains in domains poorly covered by training data.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Ontology-coupled agents significantly outperform ungrounded agents on accuracy and role consistency across enterprise domains.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"020101de2a495f90f8a8f299e84bb5f16df4ed25cbcca04ade26e9341c5b92f5"},"source":{"id":"2604.00555","kind":"arxiv","version":4},"verdict":{"id":"f7080c11-f360-4a8d-800e-5cb4286f9414","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T22:54:02.993627Z","strongest_claim":"ontology-coupled agents significantly outperform ungrounded agents on Metric Accuracy (p < .001) and Role Consistency (p < .001) across all three models with large effect sizes (Kendall's W = .46-.64)","one_line_summary":"Ontology grounding improves accuracy and role consistency of enterprise LLM agents, with larger gains in domains poorly covered by training data.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The ontologies are correctly specified and complete for the tested domains, and the controlled experiment adequately isolates the effect of ontological coupling from other factors such as prompt engineering or tool selection.","pith_extraction_headline":"Ontology-coupled agents significantly outperform ungrounded agents on accuracy and role consistency across enterprise domains."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.00555/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"d0b2adfe0366ab7325730272a88074dc0d4a33a4ba9056ad8ec319e893320da4"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}