渐进式代码切换作为大语言模型推理时的跨语言表征对齐
Gradual Code-Switching as Inference-Time Cross-Lingual Representational Alignment for LLMs
AI总结：
针对大语言模型跨语言性能不均衡问题，提出推理时的代码切换上下文学习机制，经多模型、多语言、多数据集验证，可显著提升目标及未见语言任务性能，尤其在低资源场景表现突出。
摘要：
尽管大语言模型（LLMs）在多语言场景中已取得显著进展，但由于LLMs通常依赖以英语为中心的潜在表征，其在不同语言间的性能仍不均衡。本研究中，我们提出了代码切换上下文学习（CSICL），这是一种用于跨语言表征对齐的推理时机制。CSICL不依赖突然的翻译枢纽，而是通过从目标语言逐渐过渡到英语来明确搭建推理轨迹，将非英语输入与以英语为中心的推理空间对齐。在4种LLMs、6个数据集和10种语言上，CSICL始终优于跨语言上下文学习基线，在目标语言和未见语言上分别取得6.0个百分点和4.8个百分点的平均提升。该改进可跨语系推广，在低资源场景中更为显著，目标语言和未见语言的提升分别达14.7个百分点和5.3个百分点。这些发现确立了代码切换作为推理时减少跨语言不对齐的稳健有效方法，推动LLMs向更公平高效的多语言系统发展。
英文摘要：
While large language models (LLMs) have achieved notable progress in multilingual settings, their performance remains uneven across languages as LLMs often rely on English-centric latent representations. In this work, we introduce code-switching in-context learning (CSICL), an inference-time mechanism for cross-lingual representational alignment. Rather than relying on an abrupt translation pivot, CSICL explicitly scaffolds the reasoning trajectory by gradually transitioning from a target language to English, aligning non-English inputs with an English-centric reasoning space. Across 4 LLMs, 6 datasets, and 10 languages, CSICL consistently outperforms cross-lingual in-context learning baselines, yielding average gains of 6.0pp and 4.8pp in target and unseen languages, respectively. The improvements generalize across language families and are even more pronounced in low-resource settings, with gains of 14.7pp in target and 5.3pp in unseen languages. These findings establish code-switching as a robust and effective approach for reducing cross-lingual misalignment during inference, moving LLMs towards more equitable and effective multilingual systems.
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。
