Uncertainty Propagation in LLM-Based Systems

TL;DR

This paper presents a systems-level framework for understanding how uncertainty propagates through complex LLM-based systems, highlighting the importance of principled uncertainty management across components and processes.

Contribution

It introduces a taxonomy of uncertainty propagation mechanisms at intra-model, system, and socio-technical levels, and outlines key research challenges.

Findings

Develops a conceptual framing for uncertainty signals.

Provides a structured taxonomy of propagation mechanisms.

Synthesizes engineering insights and identifies research challenges.

Abstract

Uncertainty in large language model (LLM)-based systems is often studied at the level of a single model output, yet deployed LLM applications are compound systems in which uncertainty is transformed and reused across model internals, workflow stages, component boundaries, persistent state, and human or organisational processes. Without principled treatment of how uncertainty is carried and reused across these boundaries, early errors can propagate and compound in ways that are difficult to detect and govern. This paper develops a systems-level account of uncertainty propagation. It introduces a conceptual framing for characterising propagated uncertainty signals, presents a structured taxonomy spanning intra-model (P1), system-level (P2), and socio-technical (P3) propagation mechanisms, synthesises cross-cutting engineering insights, and identifies five open research challenges.