AlphaSpace: Enabling Robotic Actions through Semantic Tokenization and Symbolic Reasoning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dao, Alan, Vu, Dinh Bach, Huy, Bui Quang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917968952164352
author Dao, Alan
Vu, Dinh Bach
Huy, Bui Quang
author_facet Dao, Alan
Vu, Dinh Bach
Huy, Bui Quang
contents This paper presents AlphaSpace, a novel methodology designed to enhance the spatial reasoning capabilities of language models for robotic manipulation in 3D Cartesian space. AlphaSpace employs a hierarchical semantics-based tokenization strategy that encodes spatial information at both coarse and fine-grained levels. Our approach represents objects with their attributes, positions, and height information through structured tokens, enabling precise spatial reasoning without relying on traditional vision-based embeddings. This approach enables LLMs to accurately manipulate objects by positioning them at specific (x, y, z) coordinates. Experimental results suggest that AlphaSpace demonstrates promising potential for improving manipulation tasks, achieving a total accuracy of 66.67%, compared to 37.5% for GPT-4o and 29.17% for Claude 3.5 Sonnet. These results demonstrate the potential of structured spatial encoding for manipulation tasks and warrant further exploration.
format Preprint
id arxiv_https___arxiv_org_abs_2503_18769
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AlphaSpace: Enabling Robotic Actions through Semantic Tokenization and Symbolic Reasoning
Dao, Alan
Vu, Dinh Bach
Huy, Bui Quang
Computation and Language
Robotics
This paper presents AlphaSpace, a novel methodology designed to enhance the spatial reasoning capabilities of language models for robotic manipulation in 3D Cartesian space. AlphaSpace employs a hierarchical semantics-based tokenization strategy that encodes spatial information at both coarse and fine-grained levels. Our approach represents objects with their attributes, positions, and height information through structured tokens, enabling precise spatial reasoning without relying on traditional vision-based embeddings. This approach enables LLMs to accurately manipulate objects by positioning them at specific (x, y, z) coordinates. Experimental results suggest that AlphaSpace demonstrates promising potential for improving manipulation tasks, achieving a total accuracy of 66.67%, compared to 37.5% for GPT-4o and 29.17% for Claude 3.5 Sonnet. These results demonstrate the potential of structured spatial encoding for manipulation tasks and warrant further exploration.
title AlphaSpace: Enabling Robotic Actions through Semantic Tokenization and Symbolic Reasoning
topic Computation and Language
Robotics
url https://arxiv.org/abs/2503.18769