Table 6 Distribution of labels in train, validation, and test splits.
From: A dataset for evaluating clinical research claims in large language models
Set | Evidence | Inconclusive | NEI | All |
|---|---|---|---|---|
Train | 366 | 262 | 632 | 1,260 |
Validation | 83 | 67 | 166 | 316 |
Test | 113 | 54 | 227 | 394 |
All | 562 | 383 | 1,025 | 1,970 |
% | 29 | 19 | 52 | 100 |
