Saved in:
Bibliographic Details
Main Authors: Luo, Licheng, Liang, Kaier, Xia, Yu, Cai, Mingyu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.13670
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908908044419072
author Luo, Licheng
Liang, Kaier
Xia, Yu
Cai, Mingyu
author_facet Luo, Licheng
Liang, Kaier
Xia, Yu
Cai, Mingyu
contents While Temporal Logic provides a rigorous verification framework for robotics, it typically operates on trajectory-level signals and does not natively represent the object-centric geometric relations that are central to manipulation. Spatio-Temporal Logic (SpaTiaL) overcomes this by explicitly capturing geometric spatial requirements, making it a natural formalism for manipulation-task verification. Consequently, translating natural language (NL) into verifiable SpaTiaL specifications is a critical objective. Yet, existing NL-to-Logic methods treat specifications as flat sequences, entangling nested temporal scopes with spatial relations and causing performance to degrade sharply under deep nesting. We propose NL2SpaTiaL, a framework modeling specifications as Hierarchical Logical Trees (HLT). By generating formulas as structured HLTs in a single shot, our approach decouples semantic parsing from syntactic rendering, aligning with human compositional spatial reasoning. To support this, we construct, to the best of our knowledge, the first NL-to-SpaTiaL dataset with explicit hierarchical supervision via a logic-first synthesis pipeline. Experiments with open-weight LLMs demonstrate that our HLT formulation significantly outperforms flat-generation baselines across various logical depths. These results show that explicit HLT structure is critical for scalable NL-to-SpaTiaL translation, ultimately enabling a rigorous ``generate-and-test'' paradigm for verifying candidate trajectories in language-conditioned robotics. Project website: https://sites.google.com/view/nl2spatial
format Preprint
id arxiv_https___arxiv_org_abs_2512_13670
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle NL2SpaTiaL: Generating Geometric Spatio-Temporal Logic Specifications from Natural Language for Manipulation Tasks
Luo, Licheng
Liang, Kaier
Xia, Yu
Cai, Mingyu
Robotics
While Temporal Logic provides a rigorous verification framework for robotics, it typically operates on trajectory-level signals and does not natively represent the object-centric geometric relations that are central to manipulation. Spatio-Temporal Logic (SpaTiaL) overcomes this by explicitly capturing geometric spatial requirements, making it a natural formalism for manipulation-task verification. Consequently, translating natural language (NL) into verifiable SpaTiaL specifications is a critical objective. Yet, existing NL-to-Logic methods treat specifications as flat sequences, entangling nested temporal scopes with spatial relations and causing performance to degrade sharply under deep nesting. We propose NL2SpaTiaL, a framework modeling specifications as Hierarchical Logical Trees (HLT). By generating formulas as structured HLTs in a single shot, our approach decouples semantic parsing from syntactic rendering, aligning with human compositional spatial reasoning. To support this, we construct, to the best of our knowledge, the first NL-to-SpaTiaL dataset with explicit hierarchical supervision via a logic-first synthesis pipeline. Experiments with open-weight LLMs demonstrate that our HLT formulation significantly outperforms flat-generation baselines across various logical depths. These results show that explicit HLT structure is critical for scalable NL-to-SpaTiaL translation, ultimately enabling a rigorous ``generate-and-test'' paradigm for verifying candidate trajectories in language-conditioned robotics. Project website: https://sites.google.com/view/nl2spatial
title NL2SpaTiaL: Generating Geometric Spatio-Temporal Logic Specifications from Natural Language for Manipulation Tasks
topic Robotics
url https://arxiv.org/abs/2512.13670