Progress-Think: Semantic Progress Reasoning for Vision-Language Navigation

TL;DR

Progress-Think introduces semantic progress reasoning for vision-language navigation, enabling agents to understand their advancement in multi-step instructions through a novel three-stage framework that improves accuracy and efficiency.

Contribution

It proposes a new semantic progress reasoning approach with a three-stage training framework, enhancing navigation performance without requiring expensive annotations.

Findings

Achieves state-of-the-art success rates on R2R-CE and RxR-CE datasets.

Demonstrates improved navigation consistency and efficiency.

Introduces a novel differentiable alignment for progress pretraining.

Abstract

Vision-Language Navigation requires agents to act coherently over long horizons by understanding not only local visual context but also how far they have advanced within a multi-step instruction. However, recent Vision-Language-Action models focus on direct action prediction and earlier progress methods predict numeric achievements; both overlook the monotonic co-progression property of the observation and instruction sequences. Building on this insight, Progress-Think introduces semantic progress reasoning, predicting instruction-style progress from visual observations to enable more accurate navigation. To achieve this without expensive annotations, we propose a three-stage framework. In the initial stage, Self-Aligned Progress Pretraining bootstraps a reasoning module via a novel differentiable alignment between visual history and instruction prefixes. Then, Progress-Guided Policy Pretraining injects learned progress states into the navigation context, guiding the policy toward consistent actions. Finally, Progress-Policy Co-Finetuning jointly optimizes both modules with tailored progress-aware reinforcement objectives. Experiments on R2R-CE and RxR-CE show state-of-the-art success and efficiency, demonstrating that semantic progress yields a more consistent representation of navigation advancement.