Cell-level error detection on MaTElDa cell-level
46.5F1 ScoreDebate
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| DebateModel=Claude 4 Sonnet, Avg Tok=28,5582026.06 | 46.5 | 68.5 | 44.1 | 80 | |
| Single AgentModel=Claude 4 Sonnet, Avg Tok=4,5392026.06 | 19.1 | 27.9 | 17.2 | 11 |
| Method | Links | ||||
|---|---|---|---|---|---|
| DebateModel=Claude 4 Sonnet, Avg Tok=28,5582026.06 | 46.5 | 68.5 | 44.1 | 80 | |
| Single AgentModel=Claude 4 Sonnet, Avg Tok=4,5392026.06 | 19.1 | 27.9 | 17.2 | 11 |