RecoverFly: A Failure-Aware Reinforcement Learning Post-Training Framework for Aerial Vision-Language Navigation

[2608.09467] RefineFly: Failure-Aware Post-Training for Aerial Vision-Language Navigation Skip to main content Search Submit Donate Log in Computer Science > Computer Vision and Pattern Recognition arXiv:2608.09467 (cs) [Submitted on 10 Aug 2026 ( v1 ), last revised 26 Sep 2026 (this version, v3)] Title: RefineFly: Failure-Aware Post-Training for Aerial Vision-Language Navigation Authors: Boxiong Wang , Hui Kang , Geng Sun , Jiahui Li , Chao Yu , Daxin Tian View PDF HTML (experimental) Abstract: Unmanned aerial vehicle vision-language navigation (UAV-VLN) requires agents to translate visual observations and language instructions into reliable flight actions in complex environments. Although recent end-to-end UAV vision-language-action (UAV-VLA) policies reduce reliance on separately designed perception, planning, and control modules, their behavior-cloning objectives provide limited supervision for errors arising during closed-loop execution. Reinforcement learning (RL) offers a promising solution, while limited interaction budgets, long-tailed scene distributions, and policy drift complicate sustained learning. To this end, we propose RefineFly, a failure-aware RL post-training framework for end-to-end UAV-VLA policies. Built on token-level proximal policy optimization (PPO), RefineFly maintains dynamic failure memory within a two-stage scene curriculum to sustain learning from unresolved tasks as training shifts from empirical to balanced scene sampling. Throughout this process, stage-specific reference regularization constrains policy deviation. Experiments on the TravelUAV benchmark demonstrate that RefineFly outperforms all comparison methods across all three splits, improving success rate over AerialVLA by 3.12 to 8.37 percentage points with a total rollout budget of about 30\% of the training-set size. Moreover, ablations reveal that the benefits of sustained failure learning vary across evaluation splits, while the two-stage scene curriculum improves overall performance, highlighting the importance of training distributions when learning from failures. Subjects: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI) Cite as: arXiv:2608.09467 [cs.CV] (or arXiv:2608.09467v3 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.09467 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Boxiong Wang [ view email ] [v1] Mon, 10 Aug 2026 11:37:46 UTC (582 KB) [v2] Fri, 28 Aug 2026 03:43:28 UTC (3,053 KB) [v3] Sat, 26 Sep 2026 11:10:02 UTC (3,167 KB) Full-text links: Access Paper: View PDF HTML (experimental) TeX Source view license Additional Features Audio Summary Current browse context: cs.CV < prev | next > new | recent | 2026-08 Change to browse by: cs cs.AI References & Citations SciX Google Scholar Semantic Scholar export BibTeX citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab)

Related documents

• RecoverFly: A Failure-Aware Reinforcement Learning Post-Training Framework for Aerial Vision-Language Navigation#682802
• CMRL: Collision-Aware and Memory-Enhanced Reinforcement Learning for UAV Navigation in Multi-Scale Obstacle Environments#973173
• WorldGuide: Learning Success-Failure Boundaries in Latent World Models for Vision-Language-Action Policies#1119279
• Dream to Fly: Model-Based Reinforcement Learning for Vision-Based Drone Flight#131282
• Jump-Start Reinforcement Learning with Vision-Language-Action Regularization#14077