---
title: "FILD-Nav:Vision-and-Language Navigation with Instruction Landmark Features in Continuous Environments"
authors:
  - "Chuangye Hu"
  - "Lulu Liu"
  - "Huaiwei Si"
  - "Yawen Zhao"
  - "Nan Ding"
doi: "10.24963/ijcai.2026/841"
url: "https://ijcai.org/proceedings/2026/841"
pdf: "https://ijcai.org/proceedings/2026/0841.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "7566-7574"
track: "AI and Robotics"
keywords:
  - "Robot control, planning, and execution with guarantees: Architectures connecting high-level intent and constraints to low-level trajectories"
  - "AIR: Robot control, planning, and execution with guarantees"
  - "Learning to understand, generalize, and explain actions: Robust generalization and transfer across tasks, objects, environments, embodiments, and long-horizon scenarios"
  - "Foundations of human–robot interaction and assistance: Learning and inference methods for aligning robot behavior with human instructions, demonstrations, and feedback"
---

# FILD-Nav:Vision-and-Language Navigation with Instruction Landmark Features in Continuous Environments

Chuangye Hu, Lulu Liu, Huaiwei Si, Yawen Zhao, Nan Ding

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 7566-7574. DOI: [10.24963/ijcai.2026/841](https://doi.org/10.24963/ijcai.2026/841)

## Abstract

Vision-and-language navigation (VLN) requires agents to follow natural language instructions to navigate autonomously in continuous environments. However, existing approaches often lack high-level semantic guidance in waypoint prediction and explicit language–landmark alignment in cross-modal planning. To address these limitations, we propose FILD-Nav, a vision-and-language navigation framework that integrates instruction landmark features. FILD-Nav extracts task-relevant landmarks from instructions and incorporates landmark semantics into both waypoint prediction and topological planning. Specifically, landmark-guided waypoint prediction improves waypoint relevance, while landmark-enhanced cross-modal planning enables more effective long-horizon navigation. Extensive experiments on the VLN-CE benchmark demonstrate that FILD-Nav consistently outperforms prior methods, achieving improvements of 2% in Success Rate (SR), 3% in Success weighted by Path Length (SPL), and 7% in Oracle Success Rate (OSR), particularly in unseen environments.