A new multi-metric fairness framework for clinical LLMs, applied to two large MIMIC-IV-based benchmarks, shows that context scarcity hurts fairness more than quantization does.
A.; Scheidegger, C.; and Venkatasubramanian, S
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
mFARM: Towards Multi-Faceted Fairness Assessment based on HARMs in Clinical Decision Support
A new multi-metric fairness framework for clinical LLMs, applied to two large MIMIC-IV-based benchmarks, shows that context scarcity hurts fairness more than quantization does.