大语言模型在长文本生成中的原子校准
Atomic Calibration of LLMs in Long-Form Generations
AI总结：
本研究提出原子校准方法，用于改进大语言模型在长文本生成中的校准能力，揭示置信度方法与生成过程中置信度变化的关联。
摘要：
大语言模型（LLMs）常常会出现幻觉，这对实际应用带来了重大挑战。置信度校准作为幻觉的有效指标，因此对于提高LLMs的可信度至关重要。先前的工作主要集中在短文本任务上，使用单一响应层面的评分（宏校准），这对于可能包含准确和不准确声明的长文本输出来说是不够的。在本工作中，我们系统地研究了原子校准，通过将长响应分解为原子声明，在细粒度层面评估事实性校准。我们进一步将现有置信度引导方法分为判别性和生成性类型，并提出了两种新的置信度融合策略以提高校准。我们的实验表明，LLMs在长文本生成过程中在原子层面表现出较差的校准能力。更重要的是，原子校准揭示了关于置信度方法对齐和生成过程中置信度变化的深刻模式。这为未来长文本生成中置信度估计的研究方向提供了新的视角。
英文摘要：
Large language models (LLMs) often suffer from hallucinations, posing significant challenges for real-world applications. Confidence calibration, as an effective indicator of hallucination, is thus essential to enhance the trustworthiness of LLMs. Prior work mainly focuses on short-form tasks using a single response-level score (macro calibration), which is insufficient for long-form outputs that may contain both accurate and inaccurate claims. In this work, we systematically study atomic calibration, which evaluates factuality calibration at a fine-grained level by decomposing long responses into atomic claims. We further categorize existing confidence elicitation methods into discriminative and generative types, and propose two new confidence fusion strategies to improve calibration. Our experiments demonstrate that LLMs exhibit poorer calibration at the atomic level during long-form generation. More importantly, atomic calibration uncovers insightful patterns regarding the alignment of confidence methods and the changes of confidence throughout generation. This sheds light on future research directions for confidence estimation in long-form generation.
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。
