Agentic Redux claims provably correct LLM agent executions on suitable domains via typed lambda calculus with full decision logging, demonstrated on healthcare compliance and security disclosure with supporting code.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Provably Auditable and Safe LLM Agents from Human-Authored Ontologies
Agentic Redux claims provably correct LLM agent executions on suitable domains via typed lambda calculus with full decision logging, demonstrated on healthcare compliance and security disclosure with supporting code.