对齐LLM以实现在企业应用中的多语言一致性
Aligning LLMs for Multilingual Consistency in Enterprise Applications
AI总结：
针对LLM在多语言企业应用中因英语中心预训练导致的性能差距，提出基于批次的多语言数据对齐微调策略，提升非英语准确率最多23.9%，且不影响英语性能。
摘要：
大型语言模型（LLMs）在全球企业应用中仍然不可靠，原因在于高资源语言与中/低资源语言之间存在显著的性能差距，这源于以英语为中心的预训练和内部推理偏差。这种不一致性削弱了多语言环境（如客户支持、内容审核和信息检索）中的客户体验和运营可靠性。即使采用先进的检索增强生成（RAG）系统，我们观察到非英语语言的准确率相比英语最多下降29%。我们提出了一种实用的、基于批次的微调对齐策略，利用每个训练批次中语义等价的多语言数据，直接对齐模型在不同语言间的输出。该方法将非英语准确率最多提升23.9%，且不影响英语性能、模型推理能力或检索质量。我们的方法易于实现、可扩展，并能无缝集成到现有的LLM训练与部署流程中，从而在工业界实现更稳健、更公平的多语言AI解决方案。
英文摘要：
Large language models (LLMs) remain unreliable for global enterprise applications due to substantial performance gaps between high-resource and mid/low-resource languages, driven by English-centric pretraining and internal reasoning biases. This inconsistency undermines customer experience and operational reliability in multilingual settings such as customer support, content moderation, and information retrieval. Even with advanced Retrieval-Augmented Generation (RAG) systems, we observe up to an 29% accuracy drop in non-English languages compared to English. We propose a practical, batch-wise alignment strategy for fine-tuning LLMs, leveraging semantically equivalent multilingual data in each training batch to directly align model outputs across languages. This approach improves non-English accuracy by up to 23.9% without compromising English performance, model reasoning, or retrieval quality. Our method is simple to implement, scalable, and integrates seamlessly with existing LLM training & deployment pipelines, enabling more robust and equitable multilingual AI solutions in industry.
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。
