LLM-generated global and local rubrics transform text inputs into structured formats that improve supervised learning performance on 15 EHRSHOT clinical tasks over count-based, naive LLM, and large clinical foundation model baselines.
## Constraints - Use only Python standard library plus ‘re‘, ‘json‘, ‘argparse‘, ‘pathlib‘, ‘sys‘
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
LLMs can construct powerful representations and streamline sample-efficient supervised learning
LLM-generated global and local rubrics transform text inputs into structured formats that improve supervised learning performance on 15 EHRSHOT clinical tasks over count-based, naive LLM, and large clinical foundation model baselines.