Search papers, labs, and topics across Lattice.
The paper introduces TRACE, a four-layer reference architecture for trustworthy agentic AI systems in operationally critical domains, featuring a classical-ML vs. LLM-validator split and stateful orchestration. A key component is a metrologically grounded trust-metric suite mapped to GUM/VIM/ISO 17025 standards. The framework emphasizes Model-Parsimony, quantified by the Computational Parsimony Ratio (CPR), and is demonstrated across three diverse applications.
Separating LLMs into a deliberate validation layer, rather than making them an architectural default, can improve trustworthiness and efficiency in agentic AI systems.
We introduce TRACE, a cross-domain engineering framework for trustworthy agentic AI in operationally critical domains. TRACE combines a four-layer reference architecture with an explicit classical-ML vs. LLM-validator split (L2a/L2b), a stateful orchestration-and-escalation policy (L3), and bounded human supervision (L4); a metrologically grounded trust-metric suite mapped to GUM/VIM/ISO 17025; and a Model-Parsimony principle quantified by the Computational Parsimony Ratio (CPR). Three instantiations--clinical decision support, industrial multi-domain operations, and a judicial AI assistant--transfer the samearchitecture and metrics across principally different governance contexts. The L2a/L2b separation makes the use of large language models a deliberate design decision rather than an architectural default, with parsimony quantified through CPR. TRACE introduces CPR as a first-class design principle in trustworthy-AI engineering.