# STRIDER: Navigation via Instruction-Aligned Structural Decision Space Optimization

Source: [STRIDER: Navigation via Instruction-Aligned Structural Decision Space Optimization](<https://www.sota2.com/research/paper/690988733ef7ccd74e5e59d2>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-11-05T05:03:08.928Z

Authors: Diqi He, Xuehao Gao, Hao Li, Junwei Han, Dingwen Zhang

Published: 2025-10-27T04:37:21.000Z

## Abstract

The Zero-shot Vision-and-Language Navigation in Continuous Environments (VLN-CE) task requires agents to navigate previously unseen 3D environments using natural language instructions, without any scene-specific training. A critical challenge in this setting lies in ensuring agents' actions align with both spatial structure and task intent over long-horizon execution. Existing methods often fail to achieve robust navigation due to a lack of structured decision-making and insufficient integration of feedback from previous actions. To address these challenges, we propose STRIDER (Instruction-Aligned Structural Decision Space Optimization), a novel framework that systematically optimizes the agent's decision space by integrating spatial layout priors and dynamic task feedback. Our approach introduces two key innovations: 1) a Structured Waypoint Generator that constrains the action space through spatial structure, and 2) a Task-Alignment Regulator that adjusts behavior based on task progress, ensuring semantic alignment throughout navigation. Extensive experiments on the R2R-CE and RxR-CE benchmarks demonstrate that STRIDER significantly outperforms strong SOTA across key metrics; in particular, it improves Success Rate (SR) from 29% to 35%, a relative gain of 20.7%. Such results highlight the importance of spatially constrained decision-making and feedback-guided execution in improving navigation fidelity for zero-shot VLN-CE.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Vision-and-Language Navigation on R2R-CE unseen continuous (val)](<https://www.sota2.com/research/sota/vision-and-language-navigation-on-r2r-ce-unseen-continuous-val.md>): Vision-and-Language Navigation; R2R-CE unseen continuous (val); SR: 35.

- [Vision-Language Navigation on RxR-CE (val-unseen)](<https://www.sota2.com/research/sota/vision-language-navigation-on-rxr-ce-val-unseen.md>): Vision-Language Navigation; RxR-CE (val-unseen); SR: 21.2.

- [Vision-Language Navigation on R2R-CE (val-unseen)](<https://www.sota2.com/research/sota/vision-language-navigation-on-r2r-ce-val-unseen.md>): Vision-Language Navigation; R2R-CE (val-unseen); Success Rate (SR): 35.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2511.00033>)

[PDF](<https://arxiv.org/pdf/2511.00033.pdf>)