Cascading Hallucination in Agentic RAG: The CHARM Framework for Detection and Mitigation
On the map: LLM reasoning → Hallucination detection and mitigation
also in 2 more
Abstract (arXiv)
Multi-step agentic retrieval-augmented generation (RAG) pipelines have demonstrated significant capability for complex reasoning tasks, yet remain vulnerable to a class of failure that existing hallucination detection mechanisms systematically miss: cascading hallucination, where errors introduced at early pipeline stages propagate and amplify across successive reasoning steps, producing confident but factually incorrect final outputs. To address this vulnerability, we formalize cascading hallucination as a distinct failure mode in agentic RAG systems, present a four-type taxonomy of cascade patterns, and introduce CHARM (Cascading Hallucination Aware Resolution and Mitigation), an architectural framework for detecting and interrupting error propagation in multi-step reasoning pipelines. CHARM comprises four components - stage-level fact verification, cross-stage consistency tracking, confidence propagation monitoring, and cascade resolution triggering - that operate alongside standard agentic RAG pipelines without requiring architectural replacement. We evaluate CHARM on HotpotQA, MuSiQue, 2WikiMultiHopQA, and a custom adversarial dataset across LangChain agentic pipeline configurations, achieving an 89.4% cascade detection rate with a 5.3% false positive rate and 215 ms +/- 18 ms average latency overhead per stage, achieving an error propagation reduction of 82.1%, compared to 18.5% for output-level detectors. Component ablations confirm that each detection module contributes meaningfully to overall cascade coverage. CHARM integrates with human-in-the-loop oversight frameworks to provide a complete reliability and governance stack for production agentic AI deployment.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Hallucination detection and mitigation, one of 17 directions in LLM reasoning. Builds on 33 of its 50 references that are in the map. First posted on arXiv 3 Jun 2026.
Builds on
33 of 50 references.
References · 33
- 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
- 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
- 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
- 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
- 2017On Calibration of Modern Neural Networkscited by 4,860
- 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
- 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
- 2023Let's Verify Step by Stepcited by 3,598
- 2017Billion-scale similarity search with GPUscited by 2,477
- 2023A Survey on Large Language Model based Autonomous Agentscited by 2,203
- 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,185
- 2020DeBERTa: Decoding-enhanced BERT with Disentangled Attentioncited by 2,157
21 more
- 2022Survey of Hallucination in Natural Language Generationcited by 2,056
- 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,800
- 2022Language Models (Mostly) Know What They Knowcited by 1,783
- 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,259
- 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,259
- 2022Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questionscited by 911
- 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 893
- 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 787
- 2022Ignore Previous Prompt: Attack Techniques For Language Modelscited by 760
- 2023Faith and Fate: Limits of Transformers on Compositionalitycited by 496
- 2023Ragas: Automated Evaluation of Retrieval Augmented Generationcited by 408
- 2023HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Modelscited by 329
- 2024A Comprehensive Survey of Hallucination Mitigation Techniques in Large Language Modelscited by 266
- 2023ARES: An Automated Evaluation Framework for Retrieval-Augmented Generation Systemscited by 259
- 2023Ever: Mitigating Hallucination in Large Language Models through Real-Time Verification and Rectificationcited by 38
- 2024OCR Hinders RAG: Evaluating the Cascading Impact of OCR on Retrieval-Augmented Generationcited by 24
- 2023TRAQ: Trustworthy Retrieval Augmented Question Answering via Conformal Predictioncited by 23
- 2026AgentHallu: Benchmarking Automated Hallucination Attribution of LLM-based Agentscited by 15
- 2025Systematization of Knowledge: Security and Safety in the Model Context Protocol Ecosystemcited by 14
- 2026SoK: Agentic Retrieval-Augmented Generation (RAG): Taxonomy, Architectures, Evaluation, and Research Directionscited by 7
- 2025Zero-knowledge LLM hallucination detection and mitigation through fine-grained cross-model consistencycited by 6
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
- 2026Hallucination Cascade: Analyzing Error Propagation in Multi-Agent LLM Systems
- 2025Mitigating Hallucination in Large Language Models (LLMs): An Application-Oriented Survey on RAG, Reasoning, and Agentic Systems
- 2025MetaRAG: Metamorphic Testing for Hallucination Detection in RAG Systems
- 2024Towards Detecting LLMs Hallucination via Markov Chain-based Multi-agent Debate Framework
- 2025Benchmarking LLM Faithfulness in RAG with Evolving Leaderboards
- 2026Mitigating Hallucination on Hallucination in RAG via Ensemble Voting
- 2025The Semantic Illusion: Certified Limits of Embedding-Based Hallucination Detection in RAG Systems