GeoDIVE: Geospatial Descent Informed by Value Estimation
cs.CV
Submitted: 2026-11-03
Updated: 2026-11-03
Terminology
Sources
- Qwen2.5-VL Technical Report
- Fast-then-Fine: A Two-Stage Framework with Multi-Granular Representation for Cross-Modal Retrieval in Remote Sensing
- UHR-BAT: Budget-Aware Token Compression Vision-Language model for Ultra-High-Resolution Remote Sensing
- Just Zoom In: Cross-View Geo-Localization via Autoregressive Zooming
- ColPali: Efficient Document Retrieval with Vision Language Models
- Learning Admissible Heuristics for A*: Theory and Practice
- GeoToken: Hierarchical Geolocalization of Images via Next Token Prediction
- Processing Megapixel Images with Deep Attention-Sampling Models
- Text2Earth: Unlocking Text-driven Remote Sensing Image Generation with a Global-Scale Dataset and a Foundation Model
- ZoomEarth: Active Perception for Ultra-High-Resolution Geospatial Vision-Language Tasks
- AdaGlimpse: Active Visual Exploration with Arbitrary Glimpse Position and Scale
- Sample Complexity of Learning Heuristic Functions for Greedy-Best-First and A* Search
- ZoomEye: Enhancing Multimodal LLMs with Human-Like Zooming Capabilities through Tree-Based Image Exploration
- DINOv3
- QuadTree Attention for Vision Transformers
- Learning When and Where to Zoom with Deep Reinforcement Learning
- GeoEyes: On-Demand Visual Focusing for Evidence-Grounded Understanding of Ultra-High-Resolution Remote Sensing Imagery
- Beyond Zooming: Learning Multi-Tool Visual Reasoning for Ultra-High-Resolution Remote Sensing
- Beyond Matryoshka: Revisiting Sparse Coding for Adaptive Representation
- MM-Matryoshka: Towards Budget-Elastic Visual Document Retrieval via a 2D Multimodal Matryoshka Training Framework
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