跨语言情感不一致：审计多语言语言模型以检测反向风险、方言表示和情感稳定性
Cross-Lingual Sentiment Misalignment: Auditing Multilingual Language Models for Inversion Risk, Dialectal Representation, and Affective Stability
AI总结：
本文通过控制基准框架评估多语言Transformer模型在平行孟加拉语-英语句子对上的表现，揭示了模型在情感反向、同理心不对称和方言表示中的问题，提出需引入情感稳定性指标以提升跨语言可靠性。
摘要：
近年来，多语言表示学习的进步旨在缩小高资源语言与低资源语言之间的性能差距，但其在跨语言保持情感意义方面的能力仍待探索，尤其是对如孟加拉语等少数语言。本研究通过引入受控基准框架，评估四种多语言Transformer模型在平行孟加拉语-英语句子对上的表现，按方言分类，以评估其表示稳定性。我们发现压缩模型架构表现出28.7%的“情感反向率”，根本性地将积极语义误解为负面（或反之）。因此，我们识别出一种跨语言情感偏移，称为“不对称同理心”，其中模型系统性地削弱或人为放大孟加拉语文本的情感权重，相对于其精确的英语对应物。最后，我们揭示了方言表示中的关键漏洞：区域模型中的“现代偏见”，其在处理正式孟加拉语注册时比现代口语文本的对齐错误增加57%。由于基础编码器继续作为LLM管道的安全分类器和奖励模型，跨语言可靠性成为关键问题。因此，我们呼吁在未来的跨语言基准中整合“情感稳定性”指标，以检测和惩罚极性反转，特别是在低资源设置中。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
