ariefansclub/han-domestic-task-ambiguity-benchmark-v1

Domestic Task Ambiguity Benchmark A benchmark dataset designed to analyze how ambiguity in human instructions affects task understanding in humanoid robots. Methods Instructions are annotated based on explicitness and reference clarity. Use Cases Ambiguity handling research Instruction disambiguation Limitations No multi-step task chains included. Part of Humanoid Network (HAN) License MIT