Correlation with human judgments on TabXBench Endurance 1.0 (test)
0.44Spearman's RhoTABXEVAL
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| TABXEVALapproach=two-phase: alignment + comparison2025.05 | 0.44 | 0.4 | 0.38 | 0.34 | 0.29 | |
| P-SCORE2025.05 | 0.3 | 0.27 | 0.24 | 0.31 | 0.39 | |
| MULTI-PROMPT + CoTChain-of-Thought=true2025.05 | 0.3 | 0.25 | 0.24 | 0.29 | 0.45 | |
| BLEURT2025.05 | 0.29 | 0.25 | 0.25 | 0.27 | 0.51 | |
| LLM RANKING2025.05 | 0.29 | 0.24 | 0.23 | 0.3 | 0.41 | |
| MULTI-PROMPTChain-of-Thought=false2025.05 | 0.29 | 0.24 | 0.23 | 0.3 | 0.42 | |
| LLM RUBRIC2025.05 | 0.23 | 0.16 | 0.17 | 0.28 | 0.47 | |
| ROUGE-L2025.05 | 0.21 | 0.18 | 0.4 | 0.27 | 0.53 | |
| BERTScore2025.05 | 0.19 | 0.15 | 0.13 | 0.25 | 0.57 | |
| EM2025.05 | 0.18 | 0.16 | 0.16 | 0.28 | 0.57 | |
| H-SCORE2025.05 | 0.14 | 0.11 | 0.09 | 0.28 | 0.51 | |
| chrF2025.05 | 0.12 | 0.11 | 0.08 | 0.25 | 0.59 | |
| TABEVAL2025.05 | -0.04 | -0.04 | -0.03 | 0.23 | 0.63 |