LoHoVLA: A Unified Vision-Language-Action Model for Long-Horizon Embodied Tasks

TL;DR

LoHoVLA is a unified vision-language-action framework that leverages a pretrained vision language model and hierarchical control to improve long-horizon embodied task performance in simulation.

Contribution

It introduces LoHoVLA, a novel unified model combining vision, language, and action for long-horizon tasks, with a hierarchical control mechanism and a new dataset LoHoSet.

Findings

Outperforms existing hierarchical and standard VLA models on long-horizon tasks

Demonstrates better generalization across diverse tasks

Shows significant improvements in the Ravens simulator

Abstract

Real-world embodied agents face long-horizon tasks, characterized by high-level goals demanding multi-step solutions beyond single actions. Successfully navigating these requires both high-level task planning (i.e., decomposing goals into sub-tasks) and low-level motion control (i.e., generating precise robot actions). While existing vision language action (VLA) models and hierarchical architectures offer potential in embodied tasks, the former often falter in planning, and the latter can suffer from coordination issues, both hampering performance. We introduce a new unified VLA framework for long-horizon tasks, dubbed LoHoVLA, to overcome these limitations. LoHoVLA leverages a large pretrained vision language model (VLM) as the backbone to jointly generate language and action tokens for sub-task generation and robot action prediction, respectively. This shared representation promotes better generalization across tasks. Additionally, LoHoVLA embraces a hierarchical closed-loop control mechanism to mitigate errors originating from both high-level planning and low-level control. To train LoHoVLA, we introduce LoHoSet, a dataset built on the Ravens simulator, containing 20 long-horizon tasks, each with 1,000 expert demonstrations composed of visual observations, linguistic goals, sub-tasks, and robot actions. Experimental results show that LoHoVLA significantly surpasses both hierarchical and standard VLA approaches on long-horizon embodied tasks in the Ravens simulator. These findings underscore the promise of unified architectures for advancing generalizable embodied intelligence.