BriefGPT.xyz
Apr, 2024
通过概率推理实现逻辑一致的语言模型
Towards Logically Consistent Language Models via Probabilistic Reasoning
HTML
PDF
Diego Calanzone, Stefano Teso, Antonio Vergari
TL;DR
通过引入基于原则的概率推理训练目标，本研究改进了大型语言模型的逻辑一致性和新知识推理能力。
Abstract
large language models
(LLMs) are a promising venue for
natural language understanding
and
generation tasks
. However, current LLMs are far
→