Model Space Reasoning as Search in Feedback Space for Planning Domain Generation

2026-04-09

AI Summary

Automatically generating high-quality, deployable planning domains from natural language descriptions remains an open challenge. This work proposes to reformulate model-based spatial reasoning as a heuristic search problem in a feedback space, leveraging large language models guided by symbolic feedback—such as validation results from VAL and landmark information—to dynamically refine the planning domain during generation. By integrating multiple symbolic constraints into the generation process, the approach substantially improves both the correctness and practical utility of the resulting domains, bringing them significantly closer to real-world deployment requirements across several benchmark tasks.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Search and Optimization: Heuristic Search
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

The generation of planning domains from natural language descriptions remains an open problem even with the advent of large language models and reasoning models. Recent work suggests that while LLMs have the ability to assist with domain generation, they are still far from producing high quality domains that can be deployed in practice. To this end, we investigate the ability of an agentic language model feedback framework to generate planning domains from natural language descriptions that have been augmented with a minimal amount of symbolic information. In particular, we evaluate the quality of the generated domains under various forms of symbolic feedback, including landmarks, and output from the VAL plan validator. Using these feedback mechanisms, we experiment using heuristic search over model space to optimize domain quality.

Problem

• planning domain generation
• natural language processing
• model space reasoning
• symbolic feedback
• domain quality

Innovation

• model space reasoning
• feedback-guided search
• planning domain generation
• symbolic feedback
• LLM-based planning