On a new 140-question Earth observation benchmark, the best LLM agent scores 33% accuracy with Google Earth Engine access because generated code fails to run over 58% of the time.
The table shows zero-shot performance with accuracy (%), failure rate (%), and selective accuracy (%)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Towards LLM Agents for Earth Observation
On a new 140-question Earth observation benchmark, the best LLM agent scores 33% accuracy with Google Earth Engine access because generated code fails to run over 58% of the time.