Cross-linguistic disagreement as a conflict of semantic alignment norms in multilingual AI~Linguistic Diversity as a Problem for Philosophy, Cognitive Science, and AI~

• URL: http://arxiv.org/abs/2503.04792v1
• Date: Sat, 01 Mar 2025 03:31:40 GMT
• Title: Cross-linguistic disagreement as a conflict of semantic alignment norms in multilingual AI~Linguistic Diversity as a Problem for Philosophy, Cognitive Science, and AI~
• Authors: Masaharu Mizumoto, Dat Tien Nguyen, Justin Sytsma, Mark Alfano, Yu Izumi, Koji Fujita, Nguyen Le Minh,
• Abstract summary: Cross-linguistic consistency (CL-consistency) seeks universal concepts across languages.<n>Folk-consistency, which respects language-specific semantic norms.<n>Findings challenge assumption that universal representations and cross-linguistic transfer capabilities are inherently desirable.
• Score: 0.2443066828522608
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multilingual large language models (LLMs) face an often-overlooked challenge stemming from intrinsic semantic differences across languages. Linguistic divergence can sometimes lead to cross-linguistic disagreements--disagreements purely due to semantic differences about a relevant concept. This paper identifies such disagreements as conflicts between two fundamental alignment norms in multilingual LLMs: cross-linguistic consistency (CL-consistency), which seeks universal concepts across languages, and consistency with folk judgments (Folk-consistency), which respects language-specific semantic norms. Through examining responses of conversational multilingual AIs in English and Japanese with the cases used in philosophy (cases of knowledge-how attributions), this study demonstrates that even state-of-the-art LLMs provide divergent and internally inconsistent responses. Such findings reveal a novel qualitative limitation in crosslingual knowledge transfer, or conceptual crosslingual knowledge barriers, challenging the assumption that universal representations and cross-linguistic transfer capabilities are inherently desirable. Moreover, they reveal conflicts of alignment policies of their developers, highlighting critical normative questions for LLM researchers and developers. The implications extend beyond technical alignment challenges, raising normative, moral-political, and metaphysical questions about the ideals underlying AI development--questions that are shared with philosophers and cognitive scientists but for which no one yet has definitive answers, inviting a multidisciplinary approach to balance the practical benefits of cross-linguistic consistency and respect for linguistic diversity.

Related papers

• Bias Beyond Borders: Political Ideology Evaluation and Steering in Multilingual LLMs [12.34382066368117]
  We present a large-scale multilingual evaluation of political bias spanning 50 countries and 33 languages.<n>We introduce a complementary post-hoc mitigation framework, Cross-Lingual Alignment Steering (CLAS), designed to augment existing steering methods.<n>Experiments demonstrate substantial bias reduction along both economic and social axes with minimal degradation in response quality.
  arXiv  Detail & Related papers  (2026-01-30T14:07:25Z)
• When Meanings Meet: Investigating the Emergence and Quality of Shared Concept Spaces during Multilingual Language Model Training [57.230355403478995]
  We investigate the development of language-agnostic concept spaces during pretraining of EuroLLM.<n>We find that shared concept spaces emerge early and continue to refine, but that alignment with them is language-dependent.<n>In contrast to prior work, our fine-grained manual analysis reveals that some apparent gains in translation quality reflect shifts in behavior.
  arXiv  Detail & Related papers  (2026-01-30T11:23:01Z)
• When Abundance Conceals Weakness: Knowledge Conflict in Multilingual Models [18.969784662298174]
  Large Language Models encode vast world knowledge across multiple languages, yet their internal beliefs are often unevenly distributed across linguistic spaces.<n> CLEAR decomposes conflict resolution into four progressive scenarios, from multilingual parametric elicitation to competitive multi-source induction.<n>In reasoning-intensive tasks, conflict resolution is dominated by language resource abundance, with high-resource languages exerting stronger persuasive power.
  arXiv  Detail & Related papers  (2026-01-11T19:26:59Z)
• MINDS: A Cross-cultural Dialogue Corpus for Social Norm Classification and Adherence Detection [8.551627762137928]
  We present Norm-RAG, a retrieval-augmented, agentic framework for nuanced social norm inference in multi-turn dialogues.<n>Our experiments show that Norm-RAG improves norm detection and generalization, demonstrates improved performance for culturally adaptive and socially intelligent dialogue systems.
  arXiv  Detail & Related papers  (2025-11-13T03:33:39Z)
• On the Entity-Level Alignment in Crosslingual Consistency [62.33186691736433]
  SubSub and SubInj integrate English translations of subjects into prompts across languages, leading to substantial gains in factual recall accuracy and consistency.<n>These interventions reinforce the entity representation alignment in the conceptual space through model's internal pivot-language processing.
  arXiv  Detail & Related papers  (2025-10-11T16:26:50Z)
• Think Natively: Unlocking Multilingual Reasoning with Consistency-Enhanced Reinforcement Learning [85.7304930030649]
  We propose M-Thinker, which is trained by a Language Consistency reward and a Cross-lingual Thinking Alignment reward.<n>M-Thinker achieves nearly 100% language consistency and superior performance on two multilingual benchmarks.
  arXiv  Detail & Related papers  (2025-10-08T17:55:02Z)
• Fairness of Automatic Speech Recognition: Looking Through a Philosophical Lens [0.42970700836450487]
  We argue that systematic misrecognition of certain speech varieties constitutes more than a technical limitation.<n>We identify three unique ethical dimensions of speech technologies that differentiate ASR bias from other algorithmic fairness concerns.
  arXiv  Detail & Related papers  (2025-08-10T02:26:47Z)
• Are Knowledge and Reference in Multilingual Language Models Cross-Lingually Consistent? [28.76156047784995]
  Cross-lingual consistency should be considered to assess cross-lingual transferability.<n>Code-switching training and cross-lingual word alignment objectives show the most promising results.
  arXiv  Detail & Related papers  (2025-07-17T06:55:15Z)
• Consistency in Language Models: Current Landscape, Challenges, and Future Directions [8.342499446600268]
  State-of-the-art language models struggle to maintain reliable consistency across different scenarios. This paper examines the landscape of consistency research in AI language systems.
  arXiv  Detail & Related papers  (2025-05-01T03:25:25Z)
• Uncovering inequalities in new knowledge learning by large language models across different languages [66.687369838071]
  We show that low-resource languages consistently face disadvantages across all four dimensions. We aim to raise awareness of linguistic inequalities in LLMs' new knowledge learning, fostering the development of more inclusive and equitable future LLMs.
  arXiv  Detail & Related papers  (2025-03-06T03:41:47Z)
• The Shrinking Landscape of Linguistic Diversity in the Age of Large Language Models [7.811355338367627]
  We show that the widespread adoption of large language models (LLMs) as writing assistants is linked to notable declines in linguistic diversity. We show that while the core content of texts is retained when LLMs polish and rewrite texts, not only do they homogenize writing styles, but they also alter stylistic elements in a way that selectively amplifies certain dominant characteristics or biases while suppressing others.
  arXiv  Detail & Related papers  (2025-02-16T20:51:07Z)
• From No to Know: Taxonomy, Challenges, and Opportunities for Negation Understanding in Multimodal Foundation Models [48.68342037881584]
  Negation, a linguistic construct conveying absence, denial, or contradiction, poses significant challenges for multilingual multimodal foundation models.<n>We propose a comprehensive taxonomy of negation constructs, illustrating how structural, semantic, and cultural factors influence multimodal foundation models.<n>We advocate for specialized benchmarks, language-specific tokenization, fine-grained attention mechanisms, and advanced multimodal architectures.
  arXiv  Detail & Related papers  (2025-02-10T16:55:13Z)
• Benchmarking Linguistic Diversity of Large Language Models [14.824871604671467]
  This paper emphasizes the importance of examining the preservation of human linguistic richness by language models. We propose a comprehensive framework for evaluating LLMs from various linguistic diversity perspectives.
  arXiv  Detail & Related papers  (2024-12-13T16:46:03Z)
• Large Language Models Reflect the Ideology of their Creators [71.65505524599888]
  Large language models (LLMs) are trained on vast amounts of data to generate natural language. This paper shows that the ideological stance of an LLM appears to reflect the worldview of its creators.
  arXiv  Detail & Related papers  (2024-10-24T04:02:30Z)
• Language Model Alignment in Multilingual Trolley Problems [138.5684081822807]
  Building on the Moral Machine experiment, we develop a cross-lingual corpus of moral dilemma vignettes in over 100 languages called MultiTP.<n>Our analysis explores the alignment of 19 different LLMs with human judgments, capturing preferences across six moral dimensions.<n>We discover significant variance in alignment across languages, challenging the assumption of uniform moral reasoning in AI systems.
  arXiv  Detail & Related papers  (2024-07-02T14:02:53Z)
• Are Structural Concepts Universal in Transformer Language Models? Towards Interpretable Cross-Lingual Generalization [27.368684663279463]
  We investigate the potential for explicitly aligning conceptual correspondence between languages to enhance cross-lingual generalization. Using the syntactic aspect of language as a testbed, our analyses of 43 languages reveal a high degree of alignability. We propose a meta-learning-based method to learn to align conceptual spaces of different languages.
  arXiv  Detail & Related papers  (2023-10-19T14:50:51Z)
• Re-Reading Improves Reasoning in Large Language Models [87.46256176508376]
  We introduce a simple, yet general and effective prompting method, Re2, to enhance the reasoning capabilities of off-the-shelf Large Language Models (LLMs) Unlike most thought-eliciting prompting methods, such as Chain-of-Thought (CoT), Re2 shifts the focus to the input by processing questions twice, thereby enhancing the understanding process. We evaluate Re2 on extensive reasoning benchmarks across 14 datasets, spanning 112 experiments, to validate its effectiveness and generality.
  arXiv  Detail & Related papers  (2023-09-12T14:36:23Z)
• NormSAGE: Multi-Lingual Multi-Cultural Norm Discovery from Conversations On-the-Fly [61.77957329364812]
  We introduce a framework for addressing the novel task of conversation-grounded multi-lingual, multi-cultural norm discovery. NormSAGE elicits knowledge about norms through directed questions representing the norm discovery task and conversation context. It further addresses the risk of language model hallucination with a self-verification mechanism ensuring that the norms discovered are correct.
  arXiv  Detail & Related papers  (2022-10-16T18:30:05Z)
• AM2iCo: Evaluating Word Meaning in Context across Low-ResourceLanguages with Adversarial Examples [51.048234591165155]
  We present AM2iCo, Adversarial and Multilingual Meaning in Context. It aims to faithfully assess the ability of state-of-the-art (SotA) representation models to understand the identity of word meaning in cross-lingual contexts. Results reveal that current SotA pretrained encoders substantially lag behind human performance.
  arXiv  Detail & Related papers  (2021-04-17T20:23:45Z)

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.