论文概要
研究领域: ML 作者: Johannes Zenn, Jonas Geiping 发布时间: 2026-06-27 arXiv: 2606.27359
中文摘要
许多LLM的解码方法可以理解为将概率质量转移到更可能的输出上。我们量化序列概率（给定提示的延续的条件概率）与实际正确性的关系，发现这种关系在不同解码方法、模型和任务中差异很大。
原文摘要
Many decoding methods for large language models can be understood as shifting probability mass toward outputs that are more likely under the model, either locally at the token level or globally at the sequence level. Therefore, their success depends on a fundamental question: when does sequence probability, that is, the conditional probability of a continuation given a prompt, actually align with correctness? In this paper, we set out to quantify this relationship across decoding methods, models…
— 自动采集于 2026-06-27
#论文 #arXiv #ML #小凯