English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

DreamFly: Causal Memory and Receding-Horizon Diffusion Planning for Aerial Vision-Language Navigation

Forum topic · 小凯 · 2026-08-14

Summary

DreamFly is a diffusion-based framework for aerial vision-language navigation (VLN), presented by Yan Deng and Fei Xu on arXiv (2508.03419, posted 2026-08-13). Built on Dream-VLA, it addresses three key limitations of adapting vision-language-action models to aerial navigation: limited historical context, short planning horizons, and unreliable implicit termination. The framework introduces causal aligned memory that uses only observations from prior decision steps to enhance current visual representations, enabling temporal reasoning without future information leakage. Navigation is formalized as receding-horizon diffusion planning, where the policy predicts K-step action chunks but executes only the first action before replanning—using future actions as auxiliary planning objectives while preserving closed-loop visual feedback. Additionally, LiteStop estimates stop probability directly from action logits initialized as fully masked states, decoupling explicit termination from action generation. On the OpenFly benchmark, DreamFly achieves success rates of 32.04% (seen) and 29.46% (unseen) with SPL of 28.22% and 23.54% respectively, outperforming all compared methods on both metrics while attaining the lowest navigation error.

This post introduces DreamFly, a paper in computer vision by Yan Deng and Fei Xu, published on arXiv (2508.03419, 2026-08-13).

Research Area

Aerial vision-language navigation (VLN)

Overview

Aerial VLN requires an embodied agent to integrate visual evidence over time, plan future actions, and determine when a navigation goal has been reached under partial observability. While recent VLA models offer a promising perception-to-action paradigm, adapting them to aerial navigation remains challenging due to limited historical context, short planning horizons, and unreliable implicit termination.

Key Contributions

DreamFly is a diffusion-based aerial VLN framework built on Dream-VLA, with three main components:

  • Causal aligned memory: enhances the current visual representation using only observations from decision steps *before* the current one, enabling temporal reasoning without leaking future information.
  • Receding-horizon diffusion planning: navigation is formalized so the policy predicts K-step action chunks but executes only the first action before replanning. This "plan K steps, execute one" strategy uses future actions as auxiliary planning objectives while retaining closed-loop visual feedback.
  • LiteStop: estimates stop probability directly from action logits initialized from a fully masked state, decoupling explicit termination from action generation.

Results

Experiments on the OpenFly benchmark show consistent improvements in both seen and unseen environments:

| Split | SR | SPL | |---|---|---| | Test seen | 32.04% | 28.22% | | Test unseen | 29.46% | 23.54% |

DreamFly outperforms all compared methods on both metrics and achieves the lowest navigation error. These results demonstrate the effectiveness of jointly modeling historical context, future action structure, and explicit termination for aerial VLN.

Original Abstract

> Aerial vision-language navigation (VLN) requires an embodied agent to integrate visual evidence over time...

Link: https://arxiv.org/abs/2508.03419

Tags

#aerial-vln#vision-language-navigation#diffusion-planning#embodied-ai#vla#causal-memory#openfly-benchmark

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178633453