When AI Writes, Whose Voice Remains? Quantifying Cultural Marker Erasure Across World English Varieties in Large Language Models

• URL: http://arxiv.org/abs/2602.22145v1
• Date: Wed, 25 Feb 2026 17:54:42 GMT
• Title: When AI Writes, Whose Voice Remains? Quantifying Cultural Marker Erasure Across World English Varieties in Large Language Models
• Abstract summary: "Cultural Ghosting" is the systematic erasure of linguistic markers unique to non-native English varieties during text processing.<n>We quantify this phenomenon using two novel metrics: Identity Erasure Rate (IER) & Semantic Preservation Score (SPS)<n>Our experiments demonstrate that explicit cultural-preservation prompts reduce erasure by 29% without sacrificing semantic quality.
• Score: 6.821769033209393
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) are increasingly used to ``professionalize'' workplace communication, often at the cost of linguistic identity. We introduce "Cultural Ghosting", the systematic erasure of linguistic markers unique to non-native English varieties during text processing. Through analysis of 22,350 LLM outputs generated from 1,490 culturally marked texts (Indian, Singaporean,& Nigerian English) processed by five models under three prompt conditions, we quantify this phenomenon using two novel metrics: Identity Erasure Rate (IER) & Semantic Preservation Score (SPS). Across all prompts, we find an overall IER of 10.26%, with model-level variation from 3.5% to 20.5% (5.9x range). Crucially, we identify a Semantic Preservation Paradox: models maintain high semantic similarity (mean SPS = 0.748) while systematically erasing cultural markers. Pragmatic markers (politeness conventions) are 1.9x more vulnerable than lexical markers (71.5% vs. 37.1% erasure). Our experiments demonstrate that explicit cultural-preservation prompts reduce erasure by 29% without sacrificing semantic quality.

Related papers

• Evaluating Multilingual Sentence Embeddings for Translation Error Detection:An English--Greek Contrastive Study [0.0]
  This study investigates whether multilingual embedding models can distinguish correct English-Greek translations.<n>It contains 1,850 examples across ten core and five exploratory error categories.<n> Embedding models detected explicit factual and lexical changes more reliably than tense-and-aspect and pronoun-coreference errors.
  arXiv  Detail & Related papers  (2026-08-28T18:32:54Z)
• On the Limitations of Cross-Lingual Consistency in Multilingual Text-to-image Generation [79.18001266925245]
  LingT2I is a benchmark covering 10 widely used languages with 33K prompts.<n>We conduct a comprehensive cross-lingual analysis, uncovering linguistic inequality and language-dependent trade-offs.
  arXiv  Detail & Related papers  (2026-08-11T14:49:23Z)
• Benchmarking Frontier LLMs on Arabic Cultural and Sociolinguistic Knowledge: A Cross-Evaluation Framework with Human SME Ground Truth [0.34797121357690153]
  Cross-evaluation framework instantiated on two underrepresented Arabic dialect communities: Egyptian and Iraqi Arabic.<n>We contribute 103 validated prompt-rubric pairs authored and graded by native-speaker SMEs.<n>Across 1,307 judge evaluations: GPT-5.4 is the most reliable judge (MADj = 10.21 pp, Signed Error = -1.12%); four of five judges show systematic leniency (+2.01% to +6.56%)
  arXiv  Detail & Related papers  (2026-06-30T20:18:32Z)
• Characterizing Cultural Localization in AI-Generated Stories [46.04734349145654]
  We propose a method to measure the degree to which content was generated through templated localization.<n>We identify the lexical tokens that distinguish stories across nationalities and measure the similarity of the narratives that remain after removing them.<n>We characterize the cultural markers for their stereotypicality and offensiveness, finding that markers from 19 countries, mostly located in the Global South, are on average offensive.
  arXiv  Detail & Related papers  (2026-06-12T16:51:52Z)
• The Cost of Perfect English: Pragmatic Flattening and the Erasure of Authorial Voice in L2 Writing Supported by GenAI [1.8115445236898322]
  We investigate "pragmatic flattening" - the systematic erasure of culturally preferred politeness and authorial stance.<n>Findings reveal a nuanced "dimensional divergence" within the Semantic Preservation Paradox.<n>We argue that future instruction must move beyond error correction to empower multilingual writers to use GenAI for linguistic enhancement.
  arXiv  Detail & Related papers  (2026-05-13T06:26:18Z)
• The GaoYao Benchmark: A Comprehensive Framework for Evaluating Multilingual and Multicultural Abilities of Large Language Models [51.61416200800499]
  GaoYao is a comprehensive benchmark with 182.3k samples, 26 languages and 51 nations/areas.<n>First, GaoYao proposes a unified framework categorizing evaluation tasks into three cultural layers.<n>Second, we achieve native-quality expansion by leveraging experts to rigorously localize subjective benchmarks into 19 languages.<n>Third, we conduct an in-depth diagnostic analysis on 20+ flagship and compact LLMs.
  arXiv  Detail & Related papers  (2026-04-22T06:19:46Z)
• Nationality encoding in language model hidden states: Probing culturally differentiated representations in persona-conditioned academic text [4.0057643036291815]
  This study tests whether Gemma-3-4b-it encodes nationality-discriminative information in hidden states when generating academic text.<n>A corpus of 270 texts was generated from 45 prompt templates crossed with six persona conditions in a 2 x 3 design.<n>Findings extend probing methodology to a sociolinguistic attribute and have practical implications for EAP and language pedagogy.
  arXiv  Detail & Related papers  (2026-04-11T10:52:05Z)
• Mitigating Cross-Lingual Cultural Inconsistencies in LLMs via Consensus-Driven Preference Optimisation [58.01855677487771]
  We show that multilingual large language models (MLLMs) frequently exhibit inconsistent behaviour when the prompt's language changes.<n>We propose Cross-lingual Cultural Consistent Preference optimisation (C-3PO), a consensus-driven alignment framework.<n>C-3PO achieves up to a 0.10-point absolute increase in $_S$ over unaligned models, outperforming strong prompting and representation steering baselines.
  arXiv  Detail & Related papers  (2026-04-02T14:04:06Z)
• "Be My Cheese?": Cultural Nuance Benchmarking for Machine Translation in Multilingual LLMs [0.0]
  We present a large-scale human evaluation benchmark for assessing cultural localisation in machine translation.<n>We evaluate 7 multilingual large language models (LLMs) across 15 target languages with 5 native-speaker raters per language.<n>Across full-text evaluations, mean overall quality is modest (1.68/3): GPT-5 (2.10/3), Claude Sonnet 3.7 (1.97/3), and Mistral Medium 3.1 (1.84/3) form the strongest tier with fewer catastrophic failures.
  arXiv  Detail & Related papers  (2026-02-04T16:35:48Z)
• Tears or Cheers? Benchmarking LLMs via Culturally Elicited Distinct Affective Responses [28.3173238194554]
  We introduce CEDAR, a benchmark constructed entirely from scenarios capturing culturally underlinetextscElicited underlinetextscDistinct underlinetextscAffective underlinetextscResponses.<n>The resulting benchmark comprises 10,962 instances across seven languages and 14 fine-grained emotion categories, with each language including 400 multimodal and 1,166 text-only samples.
  arXiv  Detail & Related papers  (2026-01-19T13:04:26Z)
• Understanding Emotion in Discourse: Recognition Insights and Linguistic Patterns for Generation [0.36980845568339205]
  We conduct a rigorous ablation study with 10-seed evaluation for recognition.<n>For linguistic analysis, we analyze 5,286 discourse marker occurrences.<n>We find a significant association between emotion and marker positioning.
  arXiv  Detail & Related papers  (2026-01-01T02:49:44Z)
• Evaluating Modern Large Language Models on Low-Resource and Morphologically Rich Languages:A Cross-Lingual Benchmark Across Cantonese, Japanese, and Turkish [12.286855282078305]
  GPT-4o, GPT-4, Claude3.5Sonnet, LLaMA3.1, MistralLarge2, LLaMA-2Chat13B, and Mistral7BInstruct are evaluated.<n>Our benchmark spans four diverse tasks: open-domain question answering, document summarization, English-to-X translation, and culturally grounded dialogue.
  arXiv  Detail & Related papers  (2025-11-05T22:09:53Z)
• DialectGen: Benchmarking and Improving Dialect Robustness in Multimodal Generation [111.94720088481614]
  Can multimodal generative models effectively produce content given dialectal textual input?<n>We construct a new large-scale benchmark spanning six common English dialects.<n>We design a general encoder-based mitigation strategy for multimodal generative models.
  arXiv  Detail & Related papers  (2025-10-16T17:56:55Z)
• COIG-Writer: A High-Quality Dataset for Chinese Creative Writing with Thought Processes [83.84578306665976]
  Large language models exhibit systematic deficiencies in creative writing, particularly in non-English contexts.<n>We present COIG-Writer, a novel Chinese creative writing dataset that captures both diverse outputs and their underlying thought processes.
  arXiv  Detail & Related papers  (2025-10-16T15:01:19Z)
• How Well Do LLMs Imitate Human Writing Style? [2.3754840025365183]
  Large language models (LLMs) can generate fluent text, but their ability to replicate the distinctive style of a specific human author remains unclear.<n>We present a fast, training-free framework for authorship verification and style imitation analysis.<n>It achieves 97.5% accuracy on academic essays and 94.5% in cross-domain evaluation.
  arXiv  Detail & Related papers  (2025-09-29T15:34:40Z)
• MMLU-ProX: A Multilingual Benchmark for Advanced Large Language Model Evaluation [86.7047714187813]
  MMLU-ProX is a benchmark covering 29 languages, built on an English benchmark.<n>Each language version consists of 11,829 identical questions, enabling direct cross-linguistic comparisons.<n>To meet efficient evaluation needs, we provide a lite version containing 658 questions per language.
  arXiv  Detail & Related papers  (2025-03-13T15:59:20Z)
• Beyond No: Quantifying AI Over-Refusal and Emotional Attachment Boundaries [0.0]
  We present an open-source benchmark and evaluation framework for assessing emotional boundary handling in Large Language Models (LLMs)<n>We evaluated three leading LLMs on their ability to maintain appropriate emotional boundaries through pattern-matched response analysis.<n>Our framework quantifies responses across seven key patterns: direct refusal, apology, explanation, deflection, acknowledgment, boundary setting, and emotional awareness.
  arXiv  Detail & Related papers  (2025-02-20T19:09:40Z)
• NusaWrites: Constructing High-Quality Corpora for Underrepresented and Extremely Low-Resource Languages [54.808217147579036]
  We conduct a case study on Indonesian local languages. We compare the effectiveness of online scraping, human translation, and paragraph writing by native speakers in constructing datasets. Our findings demonstrate that datasets generated through paragraph writing by native speakers exhibit superior quality in terms of lexical diversity and cultural content.
  arXiv  Detail & Related papers  (2023-09-19T14:42:33Z)
• We're Afraid Language Models Aren't Modeling Ambiguity [136.8068419824318]
  Managing ambiguity is a key part of human language understanding. We characterize ambiguity in a sentence by its effect on entailment relations with another sentence. We show that a multilabel NLI model can flag political claims in the wild that are misleading due to ambiguity.
  arXiv  Detail & Related papers  (2023-04-27T17:57:58Z)
• Few-Shot Cross-lingual Transfer for Coarse-grained De-identification of Code-Mixed Clinical Texts [56.72488923420374]
  Pre-trained language models (LMs) have shown great potential for cross-lingual transfer in low-resource settings. We show the few-shot cross-lingual transfer property of LMs for named recognition (NER) and apply it to solve a low-resource and real-world challenge of code-mixed (Spanish-Catalan) clinical notes de-identification in the stroke.
  arXiv  Detail & Related papers  (2022-04-10T21:46:52Z)
• Few-shot Learning with Multilingual Language Models [66.49496434282564]
  We train multilingual autoregressive language models on a balanced corpus covering a diverse set of languages. Our largest model sets new state of the art in few-shot learning in more than 20 representative languages. We present a detailed analysis of where the model succeeds and fails, showing in particular that it enables cross-lingual in-context learning.
  arXiv  Detail & Related papers  (2021-12-20T16:52:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.