Multi-step Tabular Data Processing on Multi-step tabular data processing tasks
82.49ATSProfiliTable
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| ProfiliTableBase Model=gpt-5.22026.05 | 82.49 | 48.65 | 97.3 | 96.55 | 100 | 85 | 29,901 | 195.27 | |
| ProfiliTableBase Model=gpt-4o2026.05 | 80.19 | 45.95 | 94.59 | 78.67 | 100 | 79.88 | 30,797 | 70.4 | |
| ChatDev2.0Base Model=gpt-5.22026.05 | 70.18 | 40.54 | 83.78 | 74.42 | 86.49 | 71.08 | 117,398 | 156.74 | |
| DataGovAgentBase Model=gpt-5.22026.05 | 67 | 43.24 | 81.08 | 80 | 86.49 | 71.56 | 30,778 | 108.17 | |
| MetaGPTBase Model=gpt-5.22026.05 | 66.61 | 48.65 | 81.08 | 63.27 | 83.78 | 68.68 | 84,331 | 115.65 | |
| CAMELBase Model=gpt-5.22026.05 | 62.5 | 35.14 | 78.38 | 69.77 | 81.08 | 65.37 | 68,642 | 87.46 | |
| CleanAgentBase Model=gpt-5.22026.05 | 59.4 | 24.32 | 78.38 | 49.23 | 81.08 | 58.48 | 23,570 | 45.21 | |
| ChatDev2.0Base Model=gpt-4o2026.05 | 57.26 | 27.03 | 75.68 | 65.96 | 83.78 | 61.94 | 133,688 | 143.77 | |
| MetaGPTBase Model=gpt-4o2026.05 | 49.53 | 24.32 | 59.46 | 45.61 | 70.27 | 49.84 | 30,038 | 68.91 | |
| DataGovAgentBase Model=gpt-4o2026.05 | 47.98 | 27.03 | 72.97 | 72.5 | 78.38 | 59.77 | 29,189 | 91.52 | |
| CleanAgentBase Model=gpt-4o2026.05 | 46.03 | 21.62 | 62.16 | 44.07 | 70.27 | 48.83 | 21,562 | 20.59 | |
| CAMELBase Model=gpt-4o2026.05 | 32.64 | 16.22 | 64.86 | 41.18 | 70.27 | 45.03 | 115,116 | 94.33 |