Download

Download Full Text (196 KB)

Description

PROBLEM:

  • Current LLM-based CDSS only provide post-hoc explanations, not live transparency. 
  • Clinicians cannot observe diagnostic reasoning as it unfolds.
  • Multi-agent reasoning generates long, overwhelming token streams.

GAP:

  • No existing CDSS supports token-level, real-time explainability. 
  • Flow control and semantic interpretation are tightly coupled in current systems.
  • Prior work focuses on static summaries or rule-based rationales.

PURPOSE:

  • Objective: Design EXAID, a middleware that summarizes evolving diagnostic reasoning as tokens are streamed. 
  • Goal: Enable clinicians to monitor live logic without being overwhelmed by raw data.

Publication Date

2026

EXAIM: A Real-Time Explainability Middleware for Multi-Agent Clinical Decision Support Systems

Share

COinS