Add WAM paper: FutureNav: Unified World-Action Modeling for Vision-and-Language Navigation

[279object]

Paper

• Title: FutureNav: Unified World-Action Modeling for Vision-and-Language Navigation
• Short name: FutureNav
• arXiv ID: 2606.30367v1
• Paper: https://arxiv.org/pdf/2606.30367
• Authors: Lingfeng Zhang, Zeying Gong, Xiaoshuai Hao, Haoxiang Fu, Qiang Zhang, Mingliang Zhou, Hangjun Ye, Xiaojun Liang, Junwei Liang, Wenbo Ding
• Published: 2026-06-29
• arXiv categories: cs.RO
• Matched keywords: world-action model, world-action models

README Entry

- **FutureNav**: "FutureNav: Unified World-Action Modeling for Vision-and-Language Navigation", arXiv 2026. ![](https://img.shields.io/badge/Predictive--Latent-c026d3)
  [[📄 Paper](https://arxiv.org/pdf/2606.30367)]

Robot Decision

• Decision: accept
• Confidence: high
• Taxonomy: joint_autoregressive_predictive_latent
• Taxonomy confidence: medium
• README heading: ### Autoregressive Generation
• Badges: https://img.shields.io/badge/Predictive--Latent-c026d3

Reason

Clearly models and uses action-conditioned world dynamics for embodied vision-language navigation.

Evidence

• unified world-action modeling framework for VLN
• forward dynamics objectives for modeling state transitions
• action-conditioned state transition modeling
• future spatial-state prediction for navigation actions

Taxonomy Evidence

• LLM autoregressively generates the next navigation action
• forward dynamics regresses the pooled next spatial feature
• future generation head predicts the next spatial feature
• simultaneous world and action modeling

Human Review Checklist

• This paper belongs in the WAM survey
• README section is correct
• Badges are correct
• Short name is correct
• Paper link is correct
• Merge this PR if accepted