<mods:mods xsi:schemaLocation="http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-3.xsd" version="3.3" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:mods="http://www.loc.gov/mods/v3"><mods:titleInfo><mods:title>RECREATING DIALECT FEATURES IN LITERARY TRANSLATION WITH LARGE LANGUAGE MODELS: PROBLEMS OF NEUTRALIZATION AND COMPENSATION</mods:title></mods:titleInfo><mods:name type="personal"><mods:namePart type="given">Abduvohid</mods:namePart><mods:namePart type="family">Anvarov</mods:namePart><mods:role><mods:roleTerm type="text">author</mods:roleTerm></mods:role></mods:name><mods:abstract>This article provides a theoretical analysis of the problem of reproducing dialectal features in literary translation using large language models. It distinguishes between technical normalization and literary neutralization and identifies the conditions for applying compensation strategies. As a result, a procedure for documenting translation decisions, functional losses, and residual limitations is proposed. The proposed approach has not been empirically tested.</mods:abstract><mods:originInfo><mods:dateIssued encoding="iso8601">2026-01-01</mods:dateIssued></mods:originInfo><mods:genre>Article</mods:genre></mods:mods>