Factual Inconsistency Detection on CHOCOLATE (FT)
0.291Kendall's TauBard
Evaluation Results
| Method | Links | |
|---|---|---|
| Bard2023.12 | 0.291 | |
| GPT-4V2023.12 | 0.215 | |
| CHARTVE2023.12 | 0.215 | |
| LLaVA-1.5-13B2023.12 | 0.211 | |
| ChartLlama2023.12 | 0.141 | |
| DePlot + GPT-42023.12 | 0.109 | |
| QAFACTEVAL2023.12 | 0.054 | |
| SUMMAC2023.12 | 0.036 | |
| ChartAssistant-S2023.12 | 0.036 |