VecLang reformulates multiclass vector mapping from remote sensing imagery as structured text generation using a progressive vision-language framework and reinforcement learning optimization on a new 54K-image benchmark.
Look where it matters: Training-free ultra-hr remote sensing vqa via adaptive zoom search.ArXiv, abs/2511.20460
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.CV 4years
2026 4verdicts
UNVERDICTED 4roles
baseline 1polarities
baseline 1representative citing papers
GeoVista introduces a planning-driven active perception framework with global exploration plans, branch-wise local inspection, and explicit evidence tracking to achieve state-of-the-art results on ultra-high-resolution remote sensing benchmarks.
ViRGo adaptively routes visual retrieval decisions in VLMs by estimating object scale from intrinsic localization heads combined with token confidence, matching patch retrieval on small objects, attention retrieval on large ones, and global processing when unnecessary.
ActiveScope introduces Semantic Anchor Localization (SAL) and Interference-Suppressed Refinement (ISR) to address semantic bias and contextual dominance in MLLMs, reporting 96.34% accuracy on V* Bench.
citing papers explorer
-
Vector Map as Language: Toward Unified Remote Sensing Vector Mapping
VecLang reformulates multiclass vector mapping from remote sensing imagery as structured text generation using a progressive vision-language framework and reinforcement learning optimization on a new 54K-image benchmark.
-
GeoVista: Visually Grounded Active Perception for Ultra-High-Resolution Remote Sensing Understanding
GeoVista introduces a planning-driven active perception framework with global exploration plans, branch-wise local inspection, and explicit evidence tracking to achieve state-of-the-art results on ultra-high-resolution remote sensing benchmarks.
-
Look Before You Zoom: Adaptive Routing for the Resolution-Context Trade-off in Visual RAG
ViRGo adaptively routes visual retrieval decisions in VLMs by estimating object scale from intrinsic localization heads combined with token confidence, matching patch retrieval on small objects, attention retrieval on large ones, and global processing when unnecessary.
-
ActiveScope: Actively Seeking and Correcting Perception for MLLMs
ActiveScope introduces Semantic Anchor Localization (SAL) and Interference-Suppressed Refinement (ISR) to address semantic bias and contextual dominance in MLLMs, reporting 96.34% accuracy on V* Bench.