发现多语言模型在孟加拉语和英语间存在情感误判，影响低资源语言公平性。
Cross-Lingual Sentiment Misalignment: Auditing Multilingual Language Models for Inversion Risk, Dialectal Representation, and Affective Stability
- 构建跨语言情感对齐评测框架，测试四种多语言模型在方言上的表现。
- 压缩模型情感反转率达28.7%，孟加拉语情感被系统性弱化或夸大。
- 揭示模型对现代口语更敏感，正式语体错误率高出57%，适合关注公平性的研究者。
当前多语言表征学习致力于缩小高资源与低资源语言间的性能差距，但其跨语言情感意义的保持能力仍缺乏深入研究，尤其针对孟加拉语等代表性不足的语言。本研究通过控制性基准框架，评估四种多语言Transformer模型在平行孟加拉语-英语句对上的表现，按方言分层以检验其表征稳定性。结果显示，压缩模型的情感反转率高达28.7%，将正面语义误判为负面（反之亦然）。进一步发现一种“非对称共情”现象：模型系统性削弱或人为增强孟加拉语文本的情感权重，相对于其精确英文对应句。此外，暴露关键弱点——区域模型存在“现代偏差”：处理正式孟加拉语时，对齐误差比现代口语高57%。随着基础编码器持续用于大模型的安全分类器与奖励模型，跨语言可靠性成为核心关切。因此，我们主张在未来跨语言基准中引入“情感稳定性”指标，以检测并惩罚极性反转，尤其是在低资源场景下。
原文摘要 · Abstract (English)
Recent advances in multilingual representation learning aim to bridge the performance gap between high- and low-resource languages, yet their ability to preserve affective meaning across languages remains underexplored, particularly for underrepresented languages like Bengali. This research addresses cross-lingual sentiment misalignment between Bengali and English by introducing a controlled benchmarking framework evaluating four multilingual transformer models on parallel Bengali-English sentence pairs, stratified by dialect, to assess their representational stability. We demonstrate that a compressed model architecture exhibits a 28.7% "Sentiment Inversion Rate," fundamentally misinterpreting positive semantics as negative (or vice versa). Consequently, we identify a cross-lingual sentiment skew that we call "Asymmetric Empathy," where models systematically dampen or artificially amplify the affective weight of Bengali text relative to its exact English counterpart. Finally, we expose a key vulnerability regarding dialectal representation: a "Modern Bias" in the regional model, which exhibits a 57% increase in alignment error when processing the formal Bengali register compared to modern colloquial text. As foundational encoders continue to serve as safety classifiers and reward models for LLM pipelines, cross-lingual reliability becomes a critical concern. We therefore advocate for the integration of "Affective Stability" metrics into future cross-lingual benchmarks to detect and penalize polarity inversions, particularly in low-resource settings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。