ToTTo
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Data-to-Text Generation | ToTTo | BLEU52.28 | 18 | |
| Loosely controlled table-to-text generation | ToTTO Logic2Text-style (test) | BLEU52.7 | 15 | |
| Table-to-Text Generation | ToTTo Over (test) | BLEURT0.364 | 15 | |
| Table-to-Text Generation | ToTTo Non (test) | BLEURT Score0.116 | 15 | |
| Table-to-Text Generation | ToTTo All (dev) | BLEURT0.24 | 15 | |
| Table-to-Text Generation | ToTTo (test) | BLEURT Score0.24 | 15 | |
| Data-to-Text Generation | ToTTo full (test) | BLEU50.8 | 12 | |
| Loosely controlled table-to-text generation | ToTTO Logic2Text-style (dev) | BLEU46.2 | 10 | |
| Tightly controlled table-to-text generation | ToTTO official (TestN) | BLEU48.7 | 10 | |
| Cell-Level Attribution | ToTTo | Precision74.2 | 6 | |
| Column-Level Attribution | ToTTo | Precision92.7 | 6 | |
| Row-Level Attribution | ToTTo | Precision77 | 6 | |
| Cell-level attribution | ToTTo (gold set) | Precision56.89 | 6 | |
| Open-ended table-to-text generation | ToTTO Logic2Text-style (test) | BLEU0.247 | 5 | |
| Open-ended table-to-text generation | ToTTO Logic2Text-style (TestO) | BLEU28.9 | 5 | |
| Open-ended table-to-text generation | ToTTO Logic2Text-style (TestN) | BLEU20.5 | 5 | |
| Tightly controlled table-to-text generation | ToTTO official (test) | BLEU56.7 | 5 | |
| Tightly controlled table-to-text generation | ToTTO official (dev) | BLEU49 | 5 | |
| Data-to-text generation | Totto Non-overlap (dev) | BLEU Score41.5 | 5 | |
| Data-to-text generation | Totto All (dev) | BLEU Score49.2 | 5 | |
| Table-to-text generation | ToTTo (human evaluation) | TControl89 | 4 | |
| Table-to-Text Generation | ToTTo (dev) | BLEU48.95 | 4 | |
| Data-to-text generation | Totto All (test) | BLEU49.5 | 3 | |
| Table-to-text generation | ToTTo Non-Overlap (test) | BLEU41.4 | 3 |
Showing 24 of 24 rows