01:31
2026-06-21
dev.to
natural-language-processing
How I improved my fact-checker from F1 0.655 0.813 — what actually changed
A developer improved a multilingual fact-checker's F1 score from 0.655 to 0.813 by fixing a fundamental input error: the model was trained on claims alone instead of claim-evidence pairs. The XLM-RoBE…