Trace-Level Analysis of Information Contamination in Multi-Agent Systems

TL;DR

This paper investigates how uncertainty and information contamination affect multi-agent workflows, introducing a trace-based measurement framework and empirical analysis across multiple tasks and models.

Contribution

It provides a formal taxonomy of contamination types, a trace-based detection framework, and empirical insights into contamination phenomena in structured workflows.

Findings

Workflows can diverge significantly yet still produce correct answers.

Contamination manifests as silent semantic errors, behavioral detours, or structural disruptions.

Common verification methods often fail to detect contamination effectively.

Abstract

Reasoning over heterogeneous artifacts (PDFs, spreadsheets, slide decks, etc.) increasingly occurs within structured agent workflows that iteratively extract, transform, and reference external information. In these workflows, uncertainty is not merely an input-quality issue: it can redirect decomposition and routing decisions, reshape intermediate state, and produce qualitatively different execution trajectories. We study this phenomenon by treating uncertainty as a controlled variable: we inject structured perturbations into artifact-derived representations, execute fixed workflows under comprehensive logging, and quantify contamination via trace divergence in plans, tool invocations, and intermediate state. Across 614 paired runs on 32 GAIA tasks with three different language models, we find a decoupling: workflows may diverge substantially yet recover correct answers, or remain structurally similar while producing incorrect outputs. We characterize three manifestation types: silent semantic corruption, behavioral detours with recovery, and combined structural disruption and their control-flow signatures (rerouting, extended execution, early termination). We measure operational costs and characterize why commonly used verification guardrails fail to intercept contamination. We contribute (i) a formal taxonomy of contamination manifestations in structured workflows, (ii) a trace-based measurement framework for detecting and localizing contamination across agent interactions, and (iii) empirical evidence with implications for targeted verification, defensive design, and cost control.