Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Memory-Augmented Scene Understanding and Exploration for Open-World Aerial Object-Goal Navigation
Jiacong Zhou, Jiaxu Miao, Yourun Lin, Xianyun Wang, Jun Xiao, Jun Yu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Adaptive Octree Memory Representation (ablation) | OpenVLA-based baseline · UAV-ON test (total) | SR: 12.40% -> 15.70% | Improved |
| add Instruction-Guided Memory Query on top of octree memory (ablation) | OctMem-Agent · UAV-ON test (total) | OSR: 21.10% -> 29.30% | Improved |
| add Instruction-Guided Memory Query on top of octree memory (ablation) | OctMem-Agent · UAV-ON test (total) | SR: 15.70% -> 19.50% | Improved |
| use hierarchical spatial aggregation instead of uniform voxelization (ablation) | Adaptive Octree Memory · UAV-ON test (total) | OSR: 27.50% -> 29.30% | Improved |
| use hierarchical spatial aggregation instead of uniform voxelization (ablation) | Adaptive Octree Memory · UAV-ON test (total) | SR: 19.10% -> 19.50% | Improved |
| add FiLM instruction-modulated queries (ablation) | Memory query module · UAV-ON test (total) | OSR: 27.30 -> 29.30 | Improved |
| add FiLM instruction-modulated queries (ablation) | Memory query module · UAV-ON test (total) | SR: 18.60 -> 19.50 | Improved |
| OctMem-Agent vs Navid on SPL (unseen environments) | full method · UAV-ON unseen environments | SPL: 5.99% -> 5.15% | Hurt |
| OctMem-Agent vs Navid on SPL (total) | full method · UAV-ON test (total) | SPL: 6.44% -> 6.37% | No clear effect |
| OctMem-Agent vs Navid on OSR (total) | full method · UAV-ON test (total) | OSR: 29.10% -> 29.30% | No clear effect |
Closest papers by meaning
- SatNav: A Scalable Benchmark for Long-Horizon UAV Vision-Language Navigation from Satellite Imagery (NeurIPS 2026 · similarity 0.89)
- Towards Realistic UAV Vision-Language Navigation: Platform, Benchmark, and Methodology (ICLR 2025 · similarity 0.89)
- MemoNav: Working Memory Model for Visual Navigation (CVPR 2024 · similarity 0.88)
- Parse, Search, and Confirmation: Training-Free Aerial Vision-and-Dialog Navigation with Chain-of-Thought Reasoning and Structured Spatial Memory (CVPR 2026 · similarity 0.87)
- EfficientNav: Towards On-Device Object-Goal Navigation with Navigation Map Caching and Retrieval (NeurIPS 2025 · similarity 0.87)
- WorldVLN: Autoregressive World Action Model for Aerial Vision-Language Navigation (NeurIPS 2026 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
