Search papers, labs, and topics across Lattice.
RemoteZero removes the need for human-annotated ground-truth coordinates in geospatial reasoning by leveraging the stronger discriminative ability of MLLMs to verify if a region satisfies a query, rather than directly generating coordinates. This is achieved through a box-supervision-free framework that replaces geometric supervision with intrinsic semantic verification, enabling training on unlabeled remote sensing data. Experiments demonstrate that RemoteZero achieves competitive performance compared to strong supervised methods, showcasing the potential of self-verifying training.
Unleashing geospatial reasoning on a torrent of unlabeled remote sensing data, RemoteZero rivals supervised methods by having models verify their own reasoning, not relying on human-annotated coordinates.
Geospatial reasoning requires models to resolve complex spatial semantics and user intent into precise target locations for Earth observation. Recent progress has liberated the reasoning path from manual curation, allowing models to generate their own inference chains. Yet a final dependency remains: they are still supervised by human-annotated ground-truth coordinates. This leaves the reasoning process autonomous, but not its spatial endpoint, and prevents true self-evolution on abundant unlabeled remote sensing data. To break this bottleneck, we introduce RemoteZero, a box-supervision-free framework for geospatial reasoning. RemoteZero is motivated by a simple asymmetry: an MLLM is typically better at verifying whether a region satisfies a query than at directly generating precise coordinates. Leveraging this stronger discriminative ability, RemoteZero replaces geometric supervision with intrinsic semantic verification and enables GRPO training without box annotations. The resulting framework further supports iterative self-evolution, allowing the model to improve from unlabeled remote sensing imagery through its own verification signal. Experiments show that RemoteZero achieves competitive performance against strong supervised methods, demonstrating the potential of self-verifying training for geospatial reasoning localization.
