arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.08322cs.AIcs.CL
收藏

ReCo:基于结构因果循环神经网络(SRNN)的可靠因果链推理

ReCo: Reliable Causal Chain Reasoning via Structural Causal Recurrent Neural Networks

Huawei Cloud(华为云)

Kai Xiong, Xiao Ding, Zhongyang Li, Li Du, Bing Qin, Yi Zheng, Baoxing Huai

AI总结:

针对因果链推理中的阈值效应与场景漂移问题,提出ReCo框架,利用结构因果循环神经网络(SRNN)估计外生变量间的矛盾,显著提升了模型在中英文数据集及下游任务上的表现。

摘要:

因果链推理(CCR)是许多决策AI系统的一项核心能力,要求模型通过连接因果对来构建可靠的因果链。然而,CCR面临两个主要的传递性问题:阈值效应和场景漂移。换言之,待拼接的因果对可能存在冲突的阈值边界或场景。为解决这些问题,我们提出了一种新颖的可靠因果链推理框架(ReCo)。该框架引入外生变量来表示因果链中每个因果对的阈值和场景因素,并通过结构因果循环神经网络(SRNN)估计外生变量之间的阈值与场景矛盾。实验表明,ReCo在中英文CCR数据集上均优于一系列强基线模型。此外,通过注入由ReCo蒸馏出的可靠因果链知识,BERT在四个下游因果相关任务上的表现优于由其他类型知识增强的BERT模型。

英文摘要:

Causal chain reasoning (CCR) is an essential ability for many decision-making AI systems, which requires the model to build reliable causal chains by connecting causal pairs. However, CCR suffers from two main transitive problems: threshold effect and scene drift. In other words, the causal pairs to be spliced may have a conflicting threshold boundary or scenario. To address these issues, we propose a novel Reliable Causal chain reasoning framework~(ReCo), which introduces exogenous variables to represent the threshold and scene factors of each causal pair within the causal chain, and estimates the threshold and scene contradictions across exogenous variables via structural causal recurrent neural networks~(SRNN). Experiments show that ReCo outperforms a series of strong baselines on both Chinese and English CCR datasets. Moreover, by injecting reliable causal chain knowledge distilled by ReCo, BERT can achieve better performances on four downstream causal-related tasks than BERT models enhanced by other kinds of knowledge.

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。

↑