Skip to main content
. 2024 Jun 28;10:e2092. doi: 10.7717/peerj-cs.2092

Table 3. Analysis of F1 and EM scores.

Performance measures Training Validation Dev Test
F1 76.25% 59.78% 54.19% 55.91%
EM 62.23% 44.18% 40.4% 41.4%