Research map / Vision-language and multimodal / Multimodal LLMs
AeroVerse: UAV-Agent Benchmark Suite for Simulating, Pre-Training, Finetuning, and Evaluating Aerospace Embodied Foundation Models
Fanglong Yao, Yuanchang Yue, Youzhi Liu, Zhigang Wang, Lei Jin, Bin Zhao, Jian Zhao, Xian Sun, Kun Fu
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 |
|---|---|---|---|
| fine-tune InternVL-2.5-8B on SkyAgent instruction data with embodied chain-of-thought (SkyAgent) vs GPT-4o | SkyAgent · SkyAgent-Scene3k ShangHai | BLEU: 0.1539 -> 0.4302 | Improved |
| fine-tune on SkyAgent instruction data vs GPT-4o | SkyAgent · SkyAgent-Scene3k ShangHai | SPICE: 0.1114 -> 0.3083 | Improved |
| fine-tune on SkyAgent instruction data vs GPT-4o | SkyAgent · SkyAgent-Scene3k ShangHai | LLM-Judge-Scene: 0.6800 -> 0.5166 | Hurt |
| fine-tune on SkyAgent instruction data vs GPT-4o | SkyAgent · SkyAgent-Scene3k ShenZhen | LLM-Judge-Scene: 0.7178 -> 0.4831 | Hurt |
| fine-tune on SkyAgent instruction data vs best open 2D VLM qwen-lv-7b | SkyAgent · SkyAgent-Scene3k ShangHai | BLEU: 0.2305 -> 0.4302 | Improved |
| use a general 3D VLM (3d-llm) without UAV-specific tuning vs GPT-4o | 3d-llm · SkyAgent-Scene3k ShangHai | LLM-Judge-Scene: 0.6800 -> 0.1660 | Hurt |
| add multitask curriculum learning on top of SkyAgent (SkyAgentX) | SkyAgentX · four core tasks (scene, reasoning, navigation, planning) | average improvement: 8.52% | Improved |
Closest papers by meaning
- AirCopBench: A Benchmark for Multi-drone Collaborative Embodied Perception and Reasoning (AAAI 2026 · similarity 0.87)
- U2UData+: A Scalable Swarm UAVs Autonomous Flight Dataset for Embodied Long-horizon Tasks (AAAI 2026 · similarity 0.86)
- UAV-Flow Colosseo: A Real-World Benchmark for Flying-on-a-Word UAV Imitation Learning (NeurIPS 2025 · similarity 0.86)
- Towards Realistic UAV Vision-Language Navigation: Platform, Benchmark, and Methodology (ICLR 2025 · similarity 0.85)
- Is your VLM Sky-Ready? A Comprehensive Spatial Intelligence Benchmark for UAV Navigation (CVPR 2026 · similarity 0.85)
- PhysBench: Benchmarking and Enhancing Vision-Language Models for Physical World Understanding (ICLR 2025 · similarity 0.85)
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.
