Megadose The AI curve, compressed.

Runtime Verification of Multiple Natural Language Criteria for Agent Governance

· ArXiv · AI/CL/LG ·
VFM-4B checks 1,000 natural-language rules against a 4,000-token context in under two seconds on a single H100, the authors report.

The verifier evaluates rules in parallel, marking each as satisfied, violated or inapplicable. The finetuned model outperforms GPT-5.4 on a synthetic governance benchmark and DynaGuard-4B on trace-level failure detection in DynaBench. Its individual verdicts identify violated rules and support focused escalation to a larger reasoning judge. Pretraining on CriteriaBank’s 394,000 examples improves transfer across domains, particularly when finetuning data is scarce.

ArXiv · AI/CL/LG's note

score 5

Categories: Research