The Hallucination Snowball: Modeling Error Propagation as State Transitions in Multi-Agent LLM Pipelines

[2608.14588] The Hallucination Snowball: Modeling Error Propagation as State Transitions in Multi-Agent LLM Pipelines Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Artificial Intelligence arXiv:2608.14588 (cs) [Submitted on 22 Jun 2026] Title: The Hallucination Snowball: Modeling Error Propagation as State Transitions in Multi-Agent LLM Pipelines Authors: Prabhjot Singh , Bhushan Pawar View a PDF of the paper titled The Hallucination Snowball: Modeling Error Propagation as State Transitions in Multi-Agent LLM Pipelines, by Prabhjot Singh and 1 other authors View PDF HTML (experimental) Abstract: Sequential multi-agent LLM pipelines chain specialized agents without verification at handoffs, creating a structural flaw with measurable and severe consequences. We show that hallucinations injected at Stage 1 do not merely persist; they transform: raw numerical facts become derived computations, then narrative prose, then editorially approved conclusions. At each transformation, detectability degrades near-irreversibly. We formalize this as the hallucination snowball effect, a first-order Markov process over four states (Raw Fact $\to$ Derived $\to$ Narrative $\to$ Invisible) with empirically measured per-boundary escape probabilities of 24.6%, 48.3%, and 89.3%. Across 346 automatically injected hallucinations in a 4-agent financial analysis pipeline on FinanceBench, gpt-4o detection drops from 72.0% at Stage 1 to 50.9% at Stage 4, and 23.7% of hallucinations survive completely undetected in the final output. Even the strongest model tested (Qwen3.5-397B-A17B, 87.0% at Stage 1) faces a structural ceiling; projected Stage 4 detection is only ${\sim}$60--65%. Critically, boundary gates using identical RAG verification tools reduce hallucination survival from 58.4% to 16.2% versus end-of-pipeline checking (Cohen's $h = -0.911$, $p < 0.000001$), while end-checking alone achieves merely 2.3 pp improvement over no verification. When you verify matters more than whether you verify. Our model predicts survival for $n$-agent linear pipelines and prescribes optimal verification resource allocation: invest at $S_1{\to}S_2$ first, where 75.4% of hallucinations are still catchable, not at $S_3{\to}S_4$ where 89.3% have already escaped. Comments: 10 pages, 3 figures; accepted at the FAGEN Workshop (Failure Modes in Agentic AI), ICML 2026 Subjects: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Multiagent Systems (cs.MA) Cite as: arXiv:2608.14588 [cs.AI] (or arXiv:2608.14588v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.14588 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Prabhjot Singh [ view email ] [v1] Mon, 22 Jun 2026 06:32:00 UTC (113 KB) Full-text links: Access Paper: View a PDF of the paper titled The Hallucination Snowball: Modeling Error Propagation as State Transitions in Multi-Agent LLM Pipelines, by Prabhjot Singh and 1 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.AI < prev | next > new | recent | 2026-08 Change to browse by: cs cs.CL cs.MA References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from

Related documents

• The Hallucination Snowball: Modeling Error Propagation as State Transitions in Multi-Agent LLM Pipelines#610090
• The Decomposition Tax: LLM Pipelines Lose Up to 40 Accuracy Points at Their Own Interfaces#1117156
• FAPO: Fully Autonomous Prompt Optimization of Multi-Step LLM Pipelines#290656
• Rosetta: Automating First-Principles Performance Modeling Using Multi-Agent LLMs#1013972
• Scaling Environments and Verifiers for Software Engineering Agents#634886