EarthJev: Efficient Visual Evidence Acquisition and Decision for UHR Remote Sensing Understanding
cs.CV, cs.AI
Submitted: 2026-11-04
Updated: 2026-11-04
Code: https://github.com/Da1daidaidai/EarthJev
Terminology
Sources
- Qwen2.5-VL Technical Report
- Qwen3-VL Technical Report
- GeoEyes: On-Demand Visual Focusing for Evidence-Grounded Understanding of Ultra-High-Resolution Remote Sensing Imagery
- GeoVista: Visually Grounded Active Perception for Vision-Language Understanding of Ultra-High-Resolution Remote Sensing Images
- A Benchmark for Ultra-High-Resolution Remote Sensing MLLMs
- SkySenseGPT: A Fine-Grained Instruction Tuning Dataset and Model for Remote Sensing Vision-Language Understanding
- RS-OPSD: Reliable Privileged On-Policy-Self-Distillation for Ultra-High-Resolution Remote Sensing VQA
- Beyond Zooming: Learning Multi-Tool Visual Reasoning for Ultra-High-Resolution Remote Sensing
- Zoom-Refine: Boosting High-Resolution Multimodal Understanding via Localized Zoom and Self-Refinement
- WeaveEarth: Structured Evidence Construction and Reasoning for Training-Free UHR Remote Sensing Understanding
- Look Where It Matters: Training-Free Ultra-HR Remote Sensing VQA via Adaptive Zoom Search
- The Earth in One Gaze: Training-Free Active Focus for UHR Remote Sensing Understanding
- Visual Jev: Accurate and Efficient Decisions from Shared Visual Context
- Language Models (Mostly) Know What They Know
- RSJEV: Discriminative Remote Sensing Scene Classification with Multimodal Large Language Models
- EO-Jev: Decision Model for Earth Observation
- Intern-S1: A Scientific Multimodal Foundation Model
- LLaVA-OneVision-2: Towards Next-Generation Perceptual Intelligence
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs