OLA：代码切换交互中的输出语言对齐
OLA: Output Language Alignment in Code-Switched LLM Interactions
AI总结：
OLA研究了代码切换交互中LLM输出语言对齐问题，发现现有模型存在隐含语言期望识别偏差，通过Code-Switching Aware DPO方法显著提升对齐效果。
摘要：
代码切换，即在对话中交替使用不同语言，是多语言用户自然的行为，但对大型语言模型（LLMs）提出了根本性的挑战。当用户在提示中切换语言时，通常不会指定LLM响应的期望语言，因此LLMs必须从上下文和语用线索中推断输出语言。我们发现当前LLMs系统性地未能与这一期望对齐，即使线索对人类来说是明确的，LLMs仍会以不期望的语言回应。我们引入OLA，一个评估LLMs在代码切换交互中输出语言对齐的基准。OLA专注于韩语-英语的代码切换，并涵盖了从简单的句子内混合到指令内容不匹配的多种情况。即使前沿模型也经常误解隐含的语言期望，表现出对非英语响应的偏见。我们进一步证明这种偏见不仅限于韩语，还扩展到中文和印尼语配对。模型还表现出在回应中途切换语言和语言入侵的不稳定性。链式思维提示未能解决这些错误，表明对输出语言的语用推理能力较弱。然而，具有最小数据量（约1000个示例）的代码切换感知DPO显著减少了不一致，表明这些失败源于对齐不足，而非根本限制。我们的结果强调了在现实世界中的代码切换交互中，需要将多语言LLMs与用户的隐含期望对齐的必要性。
英文摘要：
Code-switching, alternating between languages within a conversation, is natural for multilingual users, yet poses fundamental challenges for large language models (LLMs). When a user code-switches in their prompt to an LLM, they typically do not specify the expected language of the LLM response, and thus LLMs must infer the output language from contextual and pragmatic cues. We find that current LLMs systematically fail to align with this expectation, responding in undesired languages even when cues are clear to humans. We introduce OLA, a benchmark to evaluate LLMs' Output Language Alignment in code-switched interactions. OLA focuses on Korean--English code-switching and spans simple intra-sentential mixing to instruction-content mismatches. Even frontier models frequently misinterpret implicit language expectation, exhibiting a bias toward non-English responses. We further show this bias generalizes beyond Korean to Chinese and Indonesian pairs. Models also show instability through mid-response switching and language intrusions. Chain-of-Thought prompting fails to resolve these errors, indicating weak pragmatic reasoning about output language. However, Code-Switching Aware DPO with minimal data (about 1K examples) substantially reduces misalignment, suggesting these failures stem from insufficient alignment rather than fundamental limitations. Our results highlight the need to align multilingual LLMs with users' implicit expectations in real-world code-switched interactions.
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。
