Automatic Error Detection on VARIERR
7,474Error CountEVADE
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| EVADEModel=Llama-8B, Validation=one-expl2025.11 | 7,474 | 2.99 | 4.98 | 15.3 | 21 | 16.3 | |
| EVADEModel=Llama-8B, Validation=all-llm2025.11 | 6,283 | 2.35 | 4.19 | 13.9 | 14 | 10.9 | |
| EVADEModel=Llama-8B, Validation=one-llm2025.11 | 5,951 | 2.32 | 3.97 | 13.9 | 12 | 9.3 | |
| EVADEModel=Llama-70B, Validation=one-expl2025.11 | 3,354 | 2.82 | 2.24 | 18.4 | 24 | 18.6 | |
| EVADEModel=Qwen-72B, Validation=one-expl2025.11 | 3,286 | 2.83 | 2.19 | 18.8 | 26 | 20.2 | |
| EVADEModel=Qwen-7B, Validation=one-expl2025.11 | 3,282 | 2.73 | 2.19 | 16.4 | 19 | 15.3 | |
| EVADEModel=Qwen-7B, Validation=one-llm2025.11 | 3,157 | 2.44 | 2.1 | 13.6 | 15 | 12.1 | |
| EVADEModel=Llama-70B, Validation=one-llm2025.11 | 2,931 | 2.38 | 1.95 | 22.6 | 26 | 20.2 | |
| EVADEModel=Qwen-72B, Validation=one-llm2025.11 | 2,878 | 2.45 | 1.92 | 16.7 | 21 | 16.3 | |
| EVADEModel=Llama-70B, Validation=all-llm2025.11 | 2,799 | 2.33 | 1.87 | 26.8 | 31 | 24 | |
| EVADEModel=Qwen-72B, Validation=all-llm2025.11 | 2,695 | 2.15 | 1.8 | 18.8 | 22 | 17.1 | |
| EVADEModel=Qwen-7B, Validation=all-llm2025.11 | 1,883 | 1.49 | 1.26 | 14.5 | 17 | 13.7 | |
| DatamapsModel=VariErr, Validation=self2025.11 | 1,712 | 1.5 | 2.29 | 22.8 | 23.7 | 18.3 | |
| Random BaselineModel=VariErr, Validation=random2025.11 | 1,712 | 1.5 | 2.29 | 14.7 | 14.7 | 11.4 |