论文标题
审慎的语言模型符号推理者是否超过知识？
Are Pretrained Language Models Symbolic Reasoners Over Knowledge?
论文作者
论文摘要
审慎的语言模型（PLM）如何从培训集中学习事实知识？我们研究了两个最重要的机制：推理和记忆。先前的工作试图量化PLM所学习的事实的数量，但是我们使用合成数据介绍了第一项研究培训中存在的事实与PLM学到的事实之间的因果关系。为了推理，我们表明PLM似乎学会了正确地应用一些象征性推理规则，但与其他人（包括两跳推理）斗争。进一步的分析表明，即使是学习推理规则的应用也存在缺陷。为了记忆，我们确定模式一致性（其他事实系统地支持事实）和频率是其成功的关键因素。
How can pretrained language models (PLMs) learn factual knowledge from the training set? We investigate the two most important mechanisms: reasoning and memorization. Prior work has attempted to quantify the number of facts PLMs learn, but we present, using synthetic data, the first study that investigates the causal relation between facts present in training and facts learned by the PLM. For reasoning, we show that PLMs seem to learn to apply some symbolic reasoning rules correctly but struggle with others, including two-hop reasoning. Further analysis suggests that even the application of learned reasoning rules is flawed. For memorization, we identify schema conformity (facts systematically supported by other facts) and frequency as key factors for its success.
