A Modular Multitask Reasoning Framework Integrating Spatio-temporal Models and LLMs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hettige, Kethmi Hirushini, Ji, Jiahao, Long, Cheng, Xiang, Shili, Cong, Gao, Wang, Jingyuan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916810680434688
author Hettige, Kethmi Hirushini
Ji, Jiahao
Long, Cheng
Xiang, Shili
Cong, Gao
Wang, Jingyuan
author_facet Hettige, Kethmi Hirushini
Ji, Jiahao
Long, Cheng
Xiang, Shili
Cong, Gao
Wang, Jingyuan
contents Spatio-temporal data mining plays a pivotal role in informed decision making across diverse domains. However, existing models are often restricted to narrow tasks, lacking the capacity for multi-task inference and complex long-form reasoning that require generation of in-depth, explanatory outputs. These limitations restrict their applicability to real-world, multi-faceted decision scenarios. In this work, we introduce STReason, a novel framework that integrates the reasoning strengths of large language models (LLMs) with the analytical capabilities of spatio-temporal models for multi-task inference and execution. Without requiring task-specific finetuning, STReason leverages in-context learning to decompose complex natural language queries into modular, interpretable programs, which are then systematically executed to generate both solutions and detailed rationales. To facilitate rigorous evaluation, we construct a new benchmark dataset and propose a unified evaluation framework with metrics specifically designed for long-form spatio-temporal reasoning. Experimental results show that STReason significantly outperforms advanced LLM baselines across all metrics, particularly excelling in complex, reasoning-intensive spatio-temporal scenarios. Human evaluations further validate STReason's credibility and practical utility, demonstrating its potential to reduce expert workload and broaden the applicability to real-world spatio-temporal tasks. We believe STReason provides a promising direction for developing more capable and generalizable spatio-temporal reasoning systems.
format Preprint
id arxiv_https___arxiv_org_abs_2506_20073
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Modular Multitask Reasoning Framework Integrating Spatio-temporal Models and LLMs
Hettige, Kethmi Hirushini
Ji, Jiahao
Long, Cheng
Xiang, Shili
Cong, Gao
Wang, Jingyuan
Computation and Language
Artificial Intelligence
Machine Learning
Spatio-temporal data mining plays a pivotal role in informed decision making across diverse domains. However, existing models are often restricted to narrow tasks, lacking the capacity for multi-task inference and complex long-form reasoning that require generation of in-depth, explanatory outputs. These limitations restrict their applicability to real-world, multi-faceted decision scenarios. In this work, we introduce STReason, a novel framework that integrates the reasoning strengths of large language models (LLMs) with the analytical capabilities of spatio-temporal models for multi-task inference and execution. Without requiring task-specific finetuning, STReason leverages in-context learning to decompose complex natural language queries into modular, interpretable programs, which are then systematically executed to generate both solutions and detailed rationales. To facilitate rigorous evaluation, we construct a new benchmark dataset and propose a unified evaluation framework with metrics specifically designed for long-form spatio-temporal reasoning. Experimental results show that STReason significantly outperforms advanced LLM baselines across all metrics, particularly excelling in complex, reasoning-intensive spatio-temporal scenarios. Human evaluations further validate STReason's credibility and practical utility, demonstrating its potential to reduce expert workload and broaden the applicability to real-world spatio-temporal tasks. We believe STReason provides a promising direction for developing more capable and generalizable spatio-temporal reasoning systems.
title A Modular Multitask Reasoning Framework Integrating Spatio-temporal Models and LLMs
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2506.20073