Multi-task PubMedBERT model jointly predicts recall severity (Class I/II/III) and 9-class root-cause categories from 54k FDA records, outperforming single-task baselines with reported accuracy 0.963 and rho 0.983 correlation.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
RecallRisk-BERT: A Multi-Task Framework for Post-Report Medical Device Recall Triage
Multi-task PubMedBERT model jointly predicts recall severity (Class I/II/III) and 9-class root-cause categories from 54k FDA records, outperforming single-task baselines with reported accuracy 0.963 and rho 0.983 correlation.