arXiv 2312.043502024-01-17cs.CLcs.AIcs.LG

CLadder：评估语言模型中的因果推理能力

CLadder: Assessing Causal Reasoning in Language Models

MPI for Intelligent Systems, Tübingen（马克斯·普朗克智能系统研究所，蒂宾根）；ETH Zürich（苏黎世联邦理工学院）；IIT Kharagpur（印度理工学院卡拉格普尔分校）；University of Hong Kong（香港大学）；University of Washington（华盛顿大学）

Zhijing Jin, Yuen Chen, Felix Leeb, Luigi Gresele, Ojasv Kamal, Zhiheng Lyu, Kevin Blin, Fernando Gonzalez Adauto, Max Kleiman-Weiner, Mrinmaya Sachan, Bernhard Schölkopf

备注：: NeurIPS 2023; updated with CLadder dataset v1.5

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出自然语言因果推断任务及10K样本数据集CLadder，通过oracle因果推断引擎生成关联、干预和反事实问题，并引入CausalCoT提示策略，系统评估并揭示LLMs在形式化因果推理上的不足。

摘要：

执行因果推理的能力被广泛认为是智能的核心特征。在这项工作中，我们研究大型语言模型（LLMs）是否能够连贯地推理因果关系。自然语言处理（NLP）领域的许多现有工作侧重于评估LLMs中的常识性因果推理，因此未能评估模型是否能够按照一组明确定义的形式规则进行因果推断。为了解决这一问题，我们提出了一个新的NLP任务——自然语言中的因果推断，其灵感来自Judea Pearl等人提出的“因果推断引擎”。我们构建了一个包含10K样本的大规模数据集CLadder：基于一组因果图和查询（关联性、干预性和反事实性），我们通过一个oracle因果推断引擎获得符号化问题和真实答案，然后将它们翻译成自然语言。我们在该数据集上评估了多个LLMs，并引入并评估了一种定制的思维链提示策略CausalCoT。我们表明，该任务对LLMs来说极具挑战性，并进行了深入分析，以更深入地了解LLMs的因果推理能力。我们的数据已在https://huggingface.co/datasets/causalNLP/cladder开源，代码可在https://github.com/causalNLP/cladder找到。

英文摘要：

The ability to perform causal reasoning is widely considered a core feature of intelligence. In this work, we investigate whether large language models (LLMs) can coherently reason about causality. Much of the existing work in natural language processing (NLP) focuses on evaluating commonsense causal reasoning in LLMs, thus failing to assess whether a model can perform causal inference in accordance with a set of well-defined formal rules. To address this, we propose a new NLP task, causal inference in natural language, inspired by the "causal inference engine" postulated by Judea Pearl et al. We compose a large dataset, CLadder, with 10K samples: based on a collection of causal graphs and queries (associational, interventional, and counterfactual), we obtain symbolic questions and ground-truth answers, through an oracle causal inference engine. These are then translated into natural language. We evaluate multiple LLMs on our dataset, and we introduce and evaluate a bespoke chain-of-thought prompting strategy, CausalCoT. We show that our task is highly challenging for LLMs, and we conduct an in-depth analysis to gain deeper insights into the causal reasoning abilities of LLMs. Our data is open-sourced at https://huggingface.co/datasets/causalNLP/cladder, and our code can be found at https://github.com/causalNLP/cladder.

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。