A Pilot Benchmark for NL-to-FOL Translation in Planetary Exploration

In three linesPilot benchmark for translating natural language to First-Order Logic (FOL) in planetary exploration. Dataset built from NASA mission documentation (2003-2013), manually annotated with FOL representations capturing temporal structure, agent roles, and operational dependencies. Structured predicate vocabularies provided.

Summary generated by Claude — human-verified