Article

Contamination Percolation in Multi-Agent LLM Systems: A Measurement Framework and Benchmark

2026-03-31

Abstract excerpt

Multi-agent large language model (LLM) systems are being deployed in healthcare, finance, and legal settings, yet we lack reliable methods for measuring how misinformation spreads through these networks. We introduce ContamPerc, a benchmark of 400 vignettes across 10 domains (8 clinical, 2 non-clinical), 50 domain-specific semantic markers, and a measurement pipeline that separates genuine contamination from defen...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 3ab13d09-f6b2-5fd1-bbc6-9dcdad6c5a4f
DOI: 10.22541/au.177499048.88707055/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Contamination Percolation in Multi-Agent LLM Systems: A Measurement Framework and BenchmarkDOI 10.22541/au.177499048.88707055/v1Read

Emergent Misinformation Genesis in Multi-Agent LLM Clinical PipelinesSemantic proximity Read

Beyond Benchmarks: Dynamic, Automatic And Systematic Red-Teaming Agents For Trustworthy Medical Language ModelsSemantic proximity Read

Evaluating the AI Potential as a Safety Net for Diagnosis: A Novel Benchmark of Large Language Models in Correcting Diagnostic ErrorsSemantic proximity Read

FHIR-AgentEval: A Modular Sandbox for Benchmarking Clinical LLM Agents with an Evaluation of Memory-Augmented ConfigurationsSemantic proximity Read

Large Language Models in Alzheimer’s Care: Clinical Use Cases, Safety Challenges, and Implementation PathwaysSemantic proximity Read

An auditable, retrieval-grounded agent architecture improves medical accuracy and conversational safety in language modelsSemantic proximity Read

A Biosecurity Agent for Lifecycle LLM Biosecurity AlignmentSemantic proximity Read

Inducing State Anxiety in LLM Agents Reproduces Human-Like Biases in Consumer Decision-MakingSemantic proximity Read

Architectural Hallucination Mitigation in Multi-Agent Document Intelligence: Domain-Isolated RAG and Dual-Channel Claim VerificationSemantic proximity Read

Agent Role Structure and Operating Characteristics in Large Language Model Clinical Classification: A Comparative Study of Specialist and Deliberative Multi-Agent ProtocolsSemantic proximity Read

Transformation of Large Language Models: A State of ArtSemantic proximity Read

Can LLMs Get High? A Dual-Metric Framework for Evaluating Psychedelic Simulation and Safety in Large Language ModelsSemantic proximity Read