Search papers, labs, and topics across Lattice.
This paper introduces Theoria, a verification architecture that enhances trust in AI-generated answers by transforming candidate solutions into auditable sequences of typed state transitions, each supported by explicit justifications. Theoria achieves a certification rate of 91.4% strict precision on expert problems, significantly outperforming holistic LLM judges, particularly in identifying hidden premises and fabricated citations. The method not only provides a human-readable proof trace for each certification but also demonstrates complementary strengths to existing LLM approaches, addressing critical gaps in AI reasoning reliability.
Theoria achieves a remarkable 91.4% precision in certifying AI-generated answers, revealing hidden premises that traditional LLM judges often miss.
When should an AI system's answer be trusted? Formal proof assistants offer certainty but cannot reach most of the problem distribution; scalar LLM judges offer coverage but produce opaque scores that cannot be audited after the fact and are subject to the same coherence issues as any LLM. We present Theoria, a verification architecture that closes this gap. A candidate solution is rewritten into a sequence of typed state transitions, each licensed by an explicit justification, whether that be a citation, computation, or problem-given fact, and every transition is independently auditable. The foundational invariant is completeness of change: every difference between consecutive proof states must be accounted for, so hidden premises surface as unlicensed mutations rather than passing silently. On HLE-Verified Gold (185 text-only expert problems), Theoria certifies 105 at 91.4% strict precision (Wilson 95% CI [84.5%, 95.4%]). Every certification produces a human readable proof trace in which each step can be independently challenged. Holistic LLM judges achieve comparable precision at matched coverage but fail on different problems (Jaccard 0.14-0.36), making the approaches complementary. On 95 adversarial poisoned proofs across 15 domains, structured judges catch 94.7% versus 83.2% for holistic judging (p= 0.0017). The overall 11.5 pp gap concentrates in hidden premises (90.6% vs. 62.5%, a 28 pp difference) and fabricated citations (100% vs. 90%), the error classes where the formal analysis predicts an advantage; performance is identical on arithmetic and theorem-misapplication errors, where no advantage is predicted. On GPQA Diamond (n= 65), certified precision is 97.1% (Wilson CI [85.1%, 99.5%]).
