The Last Remote Sensing Model Built by Humans? Casting Any Earth Observation Task as Recursive Self-Improvement
arXiv cs.CL · 14 Oct 2026, 03:57 AEST
Earth observation models are traditionally built by hand, requiring expert design and tuning. A new approach allows agents to autonomously build and improve these models using a recursive self-improvement method.
Key facts
- Earth observation (EO) models are still built by hand.
- Any verifiable task can be cast as a harness and a verifier, where the harness defines what the agent may use and the verifier defines what counts as success.
- On four EO benchmarks spanning optical, SAR and multispectral time-series imagery, agent-built models surpass the state of the art and keep most of that accuracy from about 1% of the training labels.
- This establishes a new paradigm for EO, in which people design harnesses and verifiers and agents build the models, so EO models improve as agents do.
Read aloud
Earth observation models are traditionally built by hand, requiring expert design and tuning. A new approach allows agents to autonomously build and improve these models using a recursive self-improvement method. Earth observation EO models are still built by hand. Any verifiable task can be cast as a harness and a verifier, where the harness defines what the agent may use and the verifier defines what counts as success. On four EO benchmarks spanning optical, SAR and multispectral time-series imagery, agent-built models surpass the state of the art and keep most of that accuracy from about one percent of the training labels. This establishes a new paradigm for EO, in which people design harnesses and verifiers and agents build the models, so EO models improve as agents do.
Sources in this cluster
- The Last Remote Sensing Model Built by Humans? Casting Any Earth Observation Task as Recursive Self-Improvement
arXiv cs.CL · 14 Oct 2026, 03:57 AEST