5 papers
Interpretability without actionability: mechanistic methods cannot correct language model errors despite near-perfect internal representations
Sanjay Basu, Sadiq Y. Patel, Parth Sheth +5
Language models encode task-relevant knowledge in internal representations that far exceeds their output performance, but whether mechanistic interpretability methods can bridge th…
The Value Sensitivity Gap: How Clinical Large Language Models Respond to Patient Preference Statements in Shared Decision-Making
Sanjay Basu
Large language models (LLMs) are entering clinical workflows as decision support tools, yet how they respond to explicit patient value statements -- the core content of shared deci…
Test-Time Learning and Inference-Time Deliberation for Efficiency-First Offline Reinforcement Learning in Care Coordination and Population Health Management
Sanjay Basu, Sadiq Y. Patel, Parth Sheth +4
Care coordination and population health management programs serve large Medicaid and safety-net populations and must be auditable, efficient, and adaptable. While clinical risk for…
Hybrid Adaptive Conformal Offline Reinforcement Learning for Fair Population Health Management
Sanjay Basu, Sadiq Y. Patel, Parth Sheth +4
Population health management programs for Medicaid populations coordinate longitudinal outreach and services (e.g., benefits navigation, behavioral health, social needs support, an…
Feasibility-Guided Fair Adaptive Offline Reinforcement Learning for Medicaid Care Management
Sanjay Basu, Sadiq Y. Patel, Parth Sheth +4
We introduce Feasibility-Guided Fair Adaptive Reinforcement Learning (FG-FARL), an offline RL procedure that calibrates per-group safety thresholds to reduce harm while equalizing…