GSA-R2R
Benchmarks
TaskDatasetSOTA resultTrendResults
GSA-R2R N-Scene (test)
57SR
37
GSA-R2R N-Basic (test)
72SR
21
GSA-R2R R-Basic (test)
79Success Rate
21
GSA-R2R Basic Instructions Non-Residential v1 (test)
57.7SR
12
GSA-R2R Basic Instructions Residential v1 (test)
69.8SR
12
GSA-R2R User Instructions Residential v1 (test)
66.1SR
12
GSA-R2R Sheldon instructions
65.5SR
10
GSA-R2R Rachel instructions
68.6SR
10
GSA-R2R Moira instructions
62.3SR
10
GSA-R2R Keith instructions
68.3SR
10
GSA-R2R Child instructions
65.5SR
10
GSA-R2R (test-n-user)
58.65SR
4
GSA-R2R N-Basic (test)
75.36SR
4