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

arXiv cs.AI / 4/13/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper tackles the open challenge of generating deployable planning domains from natural-language descriptions, noting that current LLM/reasoning approaches still produce low-quality results.
It proposes an agentic language-model feedback framework that generates planning domains using only a small amount of symbolic augmentation (e.g., landmarks and validator-derived signals).
The method evaluates domain quality under different forms of symbolic feedback, including outputs from the VAL plan validator, to determine which signals most effectively improve generations.
It applies heuristic search over the model’s “model space,” using the feedback signals to iteratively optimize planning-domain quality.
Overall, the work reframes domain generation as a search-and-feedback problem rather than a single-shot generation task.

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.