arxiv:2605.14051

SPIN: Structural LLM Planning via Iterative Navigation for Industrial Tasks

Published on May 13

· Submitted by

Dhaval Patel on May 15

IBM

Upvote

Authors:

Abstract

SPIN is a planning wrapper that combines validated DAG planning with prefix-based execution control to reduce task execution and improve plan validity in industrial LLM agent systems.

AI-generated summary

Industrial LLM agent systems often separate planning from execution, yet LLM planners frequently produce structurally invalid or unnecessarily long workflows, leading to brittle failures and avoidable tool and API cost. We propose SPIN, a planning wrapper that combines validated Directed Acyclic Graph (DAG) planning with prefix based execution control. SPIN enforces a strict DAG contract through \_validate\_plan\_text and repair prompting, producing executable plans before downstream execution, and then evaluates DAG prefixes incrementally to stop when the current prefix is sufficient to answer the query. On AssetOpsBench, across 261 scenarios, SPIN reduces executed tasks from 1061 to 623 and improves Accomplished from 0.638 to 0.706, while reducing tool calls from 11.81 to 6.82 per run. On MCP Bench, the same wrapper improves planning, grounding, and dependency related scores for both GPT OSS1 and Llama 4 Maverick.

View arXiv page View PDF Add to collection

Community

DhavalPatel

Paper submitter 3 days ago

Industrial LLM agent systems often separate planning from execution, yet LLM
planners frequently produce structurally invalid or unnecessarily long
workflows, leading to brittle failures and avoidable tool and API cost. We
propose \texttt{SPIN}, a planning wrapper that combines validated Directed
Acyclic Graph (DAG) planning with prefix based execution control. \texttt{SPIN}
enforces a strict DAG contract through \texttt{_validate_plan_text} and
repair prompting, producing executable plans before downstream execution, and
then evaluates DAG prefixes incrementally to stop when the current prefix is
sufficient to answer the query. On AssetOpsBench, across 261 scenarios,
\texttt{SPIN} reduces executed tasks from 1061 to 623 and improves
\emph{Accomplished} from 0.638 to 0.706, while reducing tool calls from 11.81
to 6.82 per run. On MCP Bench, the same wrapper improves planning, grounding,
and dependency related scores for both GPT OSS1 and Llama 4 Maverick.