Computer-use Task Execution on Celestia
24.2Success RateACuRL
Evaluation Results
| Method | Links | |
|---|---|---|
| ACuRLTraining Environment=Celestia, Iteration=Iter 32026.02 | 24.2 | |
| ACuRLTraining Environment=Impress, Iteration=Iter 32026.02 | 23.2 | |
| ACuRLTraining Environment=Celestia, Iteration=Iter 22026.02 | 23.2 | |
| ACuRLTraining Environment=Thunderbird, Iteration=Iter 32026.02 | 22.2 | |
| ACuRLTraining Environment=KAlgebra, Iteration=Iter 12026.02 | 22.2 | |
| ACuRLTraining Environment=Celestia, Iteration=Iter 12026.02 | 22.2 | |
| ACuRLTraining Environment=Impress, Iteration=Iter 22026.02 | 20.2 | |
| ACuRLTraining Environment=Thunderbird, Iteration=Iter 22026.02 | 20.2 | |
| ACuRLTraining Environment=Impress, Iteration=Iter 12026.02 | 19.2 | |
| ACuRLTraining Environment=KAlgebra, Iteration=Iter 22026.02 | 19.2 | |
| UI-TARS-1.5-7BIteration=N/A2026.02 | 17.2 | |
| ACuRLTraining Environment=Writer, Iteration=Iter 12026.02 | 17.2 | |
| ACuRLTraining Environment=Writer, Iteration=Iter 22026.02 | 17.2 | |
| ACuRLTraining Environment=Writer, Iteration=Iter 32026.02 | 17.2 | |
| ACuRLTraining Environment=Calc, Iteration=Iter 12026.02 | 16.2 | |
| ACuRLTraining Environment=Thunderbird, Iteration=Iter 12026.02 | 16.2 | |
| ACuRLTraining Environment=KAlgebra, Iteration=Iter 32026.02 | 14.2 | |
| ACuRLTraining Environment=Calc, Iteration=Iter 22026.02 | 13.2 | |
| ACuRLTraining Environment=Calc, Iteration=Iter 32026.02 | 13.1 |