Lexical-Prior-Free Planning: A Symbol-Agnostic Pipeline that Enables LLMs and LRMs to Plan under Obfuscated Interfaces
Zhendong Du*, Hanliu Wang, Kenji Hashimoto
Graduate School of Information, Production and Systems, Waseda University, Kitakyushu, 808-0135, Japan
* Corresponding Author: Zhendong Du. Email: 
Computers, Materials & Continua https://doi.org/10.32604/cmc.2025.074520
Received 13 October 2025; Accepted 16 December 2025; Published online 14 January 2026
Abstract
Planning in lexical-prior-free environments presents a fundamental challenge for evaluating whether large language models (LLMs) possess genuine structural reasoning capabilities beyond lexical memorization. When predicates and action names are replaced with semantically irrelevant random symbols while preserving logical structures, existing direct generation approaches exhibit severe performance degradation. This paper proposes a symbol-agnostic closed-loop planning pipeline that enables models to construct executable plans through systematic validation and iterative refinement. The system implements a complete generate-verify-repair cycle through six core processing components: semantic comprehension extracts structural constraints, language planner generates text plans, symbol translator performs structure-preserving mapping, consistency checker conducts static screening, Stanford Research Institute Problem Solver (STRIPS) simulator executes step-by-step validation, and VAL (Validator) provides semantic verification. A repair controller orchestrates four targeted strategies addressing typical failure patterns including first-step precondition errors and mid-segment state maintenance issues. Comprehensive evaluation on PlanBench Mystery Blocksworld demonstrates substantial improvements over baseline approaches across both language models and reasoning models. Ablation studies confirm that each architectural component contributes non-redundantly to overall effectiveness, with targeted repair providing the largest impact, followed by deep constraint extraction and step-wise validation, demonstrating that superior performance emerges from synergistic integration of these mechanisms rather than any single dominant factor. Analysis reveals distinct failure patterns between model types—language models struggle with local precondition satisfaction while reasoning models face global goal achievement challenges—yet the validation-driven mechanism successfully addresses these diverse weaknesses. A particularly noteworthy finding is the convergence of final success rates across models with varying intrinsic capabilities, suggesting that systematic validation and repair mechanisms play a more decisive role than raw model capacity in lexical-prior-free scenarios. This work establishes a rigorous evaluation framework incorporating statistical significance testing and mechanistic failure analysis, providing methodological contributions for fair assessment and practical insights into building reliable planning systems under extreme constraint conditions.
Keywords
LLM planning; PDDL; symbol obfuscation; lexical-prior-free evaluation; closed-loop verification; validation-driven repair; structural reasoning; mystery domain
Open Access
Login
Register
Submit a Paper
Propose a Special lssue
Download PDF
Downloads
Citation Tools
