Fine-tuning GPT-2 on PubMed yields syntactically plausible but factually unreliable medical vignettes and distractor suggestions that could assist, not replace, human item writers.
The international per- sonality item pool and the future of public-domain personality measures,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Training Optimus Prime, M.D.: Generating Medical Certification Items by Fine-Tuning OpenAI's gpt2 Transformer Model
Fine-tuning GPT-2 on PubMed yields syntactically plausible but factually unreliable medical vignettes and distractor suggestions that could assist, not replace, human item writers.