通过词元级信息度量分析大模型不确定性，无需标注数据即可识别幻觉和高风险生成点。
LogitScope: A Framework for Analyzing LLM Uncertainty Through Information Metrics
- 基于概率分布计算熵与方差熵，实现生成过程中的逐词元不确定性评估。
- 可检测模型在生成时的高不确定区域，有效识别潜在幻觉内容。
- 轻量高效、兼容HuggingFace模型，适合研究者与工程人员部署监控。
理解并量化大型语言模型（LLM）输出中的不确定性对可靠部署至关重要。然而，传统评估方法难以提供生成过程中各词元位置的模型置信度信息。为此，我们提出LogitScope，一种通过从概率分布中计算词元级信息度量来分析LLM不确定性的轻量级框架。通过在每一步生成中测量熵与方差熵等指标，LogitScope揭示了模型置信度模式，识别潜在幻觉，并暴露模型表现出高不确定性的决策点，且无需标注数据或语义解析。我们在多种应用场景中验证了该框架的有效性，包括不确定性量化、模型行为分析及生产环境监控。该框架具备模型无关性，通过惰性计算实现高效计算，兼容任意HuggingFace模型，使研究人员与实践者可在推理阶段便捷地检视模型行为。
原文摘要 · Abstract (English)
Understanding and quantifying uncertainty in large language model (LLM) outputs is critical for reliable deployment. However, traditional evaluation approaches provide limited insight into model confidence at individual token positions during generation. To address this issue, we introduce LogitScope, a lightweight framework for analyzing LLM uncertainty through token-level information metrics computed from probability distributions. By measuring metrics such as entropy and varentropy at each generation step, LogitScope reveals patterns in model confidence, identifies potential hallucinations, and exposes decision points where models exhibit high uncertainty, all without requiring labeled data or semantic interpretation. We demonstrate LogitScope's utility across diverse applications including uncertainty quantification, model behavior analysis, and production monitoring. The framework is model-agnostic, computationally efficient through lazy evaluation, and compatible with any HuggingFace model, enabling both researchers and practitioners to inspect LLM behavior during inference.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。