Quantum Anti-Hallucination Firewall: Multi-World Verification for RAG Systems.

Владислав Витальевич Тишков

ПРЕПРИНТ

Эта статья является препринтом и не была отрецензирована.
О результатах, изложенных в препринтах, не следует сообщать в СМИ как о проверенной информации.

Quantum Anti-Hallucination Firewall: Multi-World Verification for RAG Systems.

В. В. Тишков

2026-08-23

https://doi.org/10.24108/preprints-3116226

We present the Quantum Anti-Hallucination Firewall (QAF), a novel verification framework for detecting hallucinations in Retrieval-Augmented Generation (RAG) systems. The framework addresses a fundamental limitation identified by Gödel's incompleteness theorems: any single verification system contains statements whose truth value cannot be determined within that system. Our approach deploys multiple independent verification agents operating in parallel 'quantum worlds'—different semantic and syntactic representations of the same claim. When these agents converge on a verdict with high confidence, we identify the genomic code of truth—a unique, reproducible signature of verified information. The implementation comprises three specialized verification agents (fact checker, contradiction detector, completeness evaluator), a differential convergence equation for modeling confidence evolution, and an apogee detection algorithm for finding optimal verification points on the Pareto frontier. We prove that as the number of independent verification systems increases, the probability of a statement being undecidable across all systems asymptotically approaches zero. The framework is modular, extensible, and available as open-source software.

Ссылка для цитирования:

Тишков В. В. 2026. Quantum Anti-Hallucination Firewall: Multi-World Verification for RAG Systems. PREPRINTS.RU. https://doi.org/10.24108/preprints-3116226

Список литературы

1. J. Maynez et al., "On faithfulness and factuality in abstractive summarization," in Proceedings of ACL, 2020, pp. 1906-1919.

2. K. Shuster et al., "Retrieval augmentation reduces hallucination in conversation," in Findings of EMNLP, 2021, pp. 3784-3803.

3. J. Li et al., "HaluEval: A large-scale hallucination evaluation benchmark for large language models," in Proceedings of EMNLP, 2023.

4. P. Lewis et al., "Retrieval-augmented generation for knowledge-intensive NLP tasks," in NeurIPS, vol. 33, 2020, pp. 9459-9474.

5. Z. Guo et al., "A survey on automated fact-checking," TACL, vol. 10, pp. 178-206, 2022.

6. N. Reimers and I. Gurevych, "Sentence-BERT: Sentence embeddings using Siamese BERT-networks," in Proceedings of EMNLP-IJCNLP, 2019, pp. 3982-3992.

7. Y. Liu et al., "RoBERTa: A robustly optimized BERT pretraining approach," arXiv:1907.11692, 2019.

8. W. Kryściński et al., "Evaluating the factual consistency of abstractive text summarization," in Proceedings of EMNLP, 2020, pp. 9332-9346.

9. P. Laban et al., "SummaC: Re-visiting NLI-based models for inconsistency detection in summarization," TACL, vol. 10, pp. 163-177, 2022.

10. P. Manakul et al., "SelfCheckGPT: Zero-resource black-box hallucination detection for generative large language models," in Proceedings of EMNLP, 2023.

11. R. Cohen et al., "LM vs LM: Detecting factual errors via cross examination," in Proceedings of EMNLP, 2023.

12. Y. Du et al., "Improving factuality and reasoning in language models through multiagent debate," arXiv:2305.14325, 2023.

13. K. Gödel, "Über formal unentscheidbare Sätze der Principia Mathematica und verwandter Systeme I," Monatshefte für Mathematik und Physik, vol. 38, pp. 173-198, 1931.

14. J. Thorne et al., "FEVER: A large-scale dataset for fact extraction and verification," in Proceedings of NAACL-HLT, 2018, pp. 809-819.

15. Y. Zha et al., "AlignScore: Evaluating factual consistency with a unified alignment function," in Proceedings of ACL, 2023.

16. J. R. Lucas, "Minds, machines and Gödel," Philosophy, vol. 36, pp. 112-127, 1961.

17. R. Penrose, The Emperor's New Mind. Oxford University Press, 1989.

18. V.V. Tishkov, "A Generalisation of Bertrand's Theorem to Anisotropic Harmonic Oscillators: Geometry and Classification of Lissajous Helices," Preprints.ru, 2026. DOI: https://doi.org/10.24108/preprints-3115944

19. V.V. Tishkov, "Explicit Formula for the Symmetric Coherence Kernel Associated with Dirichlet L-Functions," Preprints.ru, 2026. DOI: https://doi.org/10.24108/preprints-3115945

20. V.V. Tishkov, "From Graph Laplacians to String Partition Functions: A Rigorous Pathway from Discrete Spectra to Emergent Geometry," arXiv:2605.00452, 2026.

ПРЕПРИНТ

ИНФОРМАЦИЯ

Использование куки-файлов