Central Asian Research Index

Polytechnic Bulletin. Intellect, Innovations, Investments Series, 2026

HALLUCINATION MITIGATION IN LLMs VIA HYBRID RAG + CLASSICAL ML VERIFICATION PIPELINE

Shavkat Kholov, K.M. Bobobekov, Husein Abdullozoda, Hasan Abdullozoda

Аннотация · Abstract

Hallucinations - factually incorrect yet confidently stated outputs - remain a critical challenge in large language models (LLMs). This paper proposes a hybrid pipeline that combines Retrieval-Augmented Generation (RAG) with a classical machinelearning verification module (SVM, logistic regression) for automatic hallucination detection and correction. The proposed architecture is evaluated on a corpus of university academic documents in English and Tajik. Experimental results show that the hybrid approach reduces the hallucination rate by 41 % compared to a plain LLM baseline and by 23 % compared to RAG without verification, achieving an F1-score of 0.90 on the English corpus and 0.85 on the Tajik corpus.

Сведения · Details