AI summaryⓘ
The authors point out that large language models (LLMs) can spot patterns but often fail to track when information is unreliable or made up. They previously developed a method called POLANYI++ to create detailed knowledge graphs from LLM outputs, but it still accepted false information without noticing. To fix this, they created DARKSIDE, which adds a system to keep track of whether statements are trustworthy, questionable, or fabricated, and can flag when too much unreliable info appears. They tested DARKSIDE on tricky, nonsense-filled questions and found it helps catch errors by using a special memory and a way to check evidence. This shows that adding explicit checks for coherence can improve how LLMs handle complex or misleading inputs.
Large Language ModelsLogic-Augmented GenerationPOLANYI++Extended Knowledge GraphCoherence AuditingOntologyWarrant AxisAdversarial TestingEpistemic Firewall
Abstract
Large Language Models (LLMs) recognise patterns but do not natively track the path of exclusions that a coherent discourse demands. When an input rests on a fabricated authority, a misapplied mechanism, or a surreptitious analogy, an unsteered LLM tends to engage with it as if it were grounded, and to reify the misstep into any structured output it generates. Logic-Augmented Generation (LAG) with POLANYI++, an LLM-steering method that uses heuristics, ontologies and problem solving methods for tacit knowledge extraction, produces an Extended Knowledge Graph (XKG) in OWL2, but inherits the same vulnerability: a sophisticated nonsensical input is reified into the graph alongside the legitimate triples, and is hardly detectable by automated reasoners since the XKG is generated jointly with the wrong assumptions. We introduce DARKSIDE, a coherence auditing method on top of POLANYI++. It formalises the trail as an explicit data structure of accumulated exclusions over discourse time, complemented by a warrant axis that classifies each named referent as Warranted, Unattested, Misattributed or Fabricated, with an escalation rule that pushes the DelegationRiskAssessment to UNSAFE when the fabricated rate is positive or the unsupported rate exceeds a threshold. We evaluate DARKSIDE as a steering layer over a Gemini 3 on BSBench, a 100-item adversarial corpus of sophisticated-sounding nonsense across software engineering, finance, healthcare, physics and law, with Claude Sonnet 4.6 as an independent judge. The empirical evidence supports an architectural claim: when an LLM forward pass is wrapped in an ontology-mediated negative-trail apparatus, the structural pattern-vs-path gap can be partially scaffolded. The XKG functions as the missing memory, and the warrant axis as an epistemic firewall.