A2A
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
A2A low-complexity subtask 2
93Success Rate
19
A2A whole (full)
87Success Rate (SR)
10
A2A high-complexity portion subtask >= 3
80Success Rate (SR)
10
A2A
87SR
9
A2A high-complexity subtask >= 3
80SR
9
a2a (test)
0.335Loss
7