Plan-Then-Verify: Verifier-First Planning for Early Feasibility Validation in LLM Agent Systems
Applied Sciences · 01.11.2026 · 16 (21) · 10828
DOI: 10.3390/app162110828 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 01.11.2026. Verfasst von Emil Benedykciuk und 6 weiteren Autor:innen. Die Arbeit nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: European Union.
Abstract (Originalsprache)
Pre-execution feasibility validation can substantially improve LLM-based task planning without proportional inference cost, but only when it is grounded, deterministic, and tightly bounded. We propose Plan-Then-Verify (PTV), a verifier-first control pattern in which a candidate workflow is checked against a documented knowledge base before execution begins. A planner generates a structured task graph, a deterministic verifier validates functions, artifacts, and ordering constraints, and one bounded repair pass applies local edits when possible. Unsupported workflows are blocked through safe abstention rather than hallucinated capability invention. We evaluate PTV on Core-60, a 60-task public benchmark spanning supported, multi-step, repair-sensitive, and unsupported requests. Across local and server-side model panels, PTV substantially improves the reported static conformance score on several model families, including Gemma, Devstral, and DeepSeek, while remaining close to Plan-and-Execute in token cost. This score combines supported-task success with correct abstention on unsupported requests; it is not a real-execution success rate. Paired bootstrap confidence intervals show positive PTV–PAE conformance differences for the positive model families and near-zero effects for model-serving configurations whose structured outputs collapse. Additional schema-constrained reruns show that this boundary condition is not merely a scoring artifact: Qwen and DeepSeek recover usable structured plans and then show clear PTV gains, whereas GPT-OSS and GLM remain fragile under the tested serving stack. The primary benchmark is static and invokes no real tools; an auxiliary sandbox replay executes stored plans against mock tools derived from the same specification. The primary outcomes measure conformance to the frozen knowledge base and feasibility criteria, while the replay provides side-effect-free trace checks. These results suggest that lightweight feasibility validation improves static SOP conformance primarily by amplifying planners capable of producing stable structured workflows, rather than compensating for weak upstream plan representations. Core-60, the knowledge-base specification, implementation, and analysis materials are publicly available.
Förderung
- European Union
Zitieren
APA
Benedykciuk, E., Denkowski, M., Jezior, M., Kawiak, A., Kobus, A., Kurasiewicz, S., & Wójcik, G. M. (2026). Plan-Then-Verify: Verifier-First Planning for Early Feasibility Validation in LLM Agent Systems. Applied Sciences, 16(21), 10828. https://doi.org/10.3390/app162110828
Vancouver
Benedykciuk E, Denkowski M, Jezior M, Kawiak A, Kobus A, Kurasiewicz S, et al. Plan-Then-Verify: Verifier-First Planning for Early Feasibility Validation in LLM Agent Systems. Applied Sciences. 2026;16(21):10828. doi:10.3390/app162110828
BibTeX
@article{Benedykciuk2026,
author = {Benedykciuk, Emil and Denkowski, Marcin and Jezior, Marek and Kawiak, Andrzej and Kobus, Adam and Kurasiewicz, Stanisław and Wójcik, Grzegorz Marcin},
title = {Plan-Then-Verify: Verifier-First Planning for Early Feasibility Validation in LLM Agent Systems},
journal = {Applied Sciences},
year = {2026},
volume = {16},
number = {21},
pages = {10828},
doi = {10.3390/app162110828}
}
EUVIMED