测试大模型对因果图的推理能力，发现编码方式严重影响结果准确率。
CausalGraph2LLM: Evaluating LLMs for Causal Queries
- 构建超70万条查询的因果推理基准，区分图级与节点级任务
- GPT-4等模型在不同编码下准确率波动达60%，表现极不稳定
- 揭示模型受上下文干扰的偏差问题，适合关注可信AI的研究者
因果关系在科学研究中至关重要，用于揭示变量间的真正关联。这些因果关系常以有向无环图（causal graphs）表示。随着大语言模型（LLMs）的发展，其在因果推理及假设因果图方面的潜力受到关注。这要求模型能有效编码因果图以支持下游任务。本文提出CausalGraph2LLM，一个涵盖超过70万条查询、多种因果图场景的综合性基准，用于评估LLMs的因果推理能力。我们将因果查询分为图级和节点级两类，并对开源与专有模型进行评测。结果显示，尽管LLMs在该领域展现潜力，但其性能高度依赖编码方式，即使强大模型如GPT-4和Gemini-1.5也表现出约60%的准确率波动。我们进一步验证了编码敏感性对下游因果干预任务的影响。此外，模型在接触因果图上下文信息时常出现偏差，可能源于其参数化记忆。
原文摘要 · Abstract (English)
Causality is essential in scientific research, enabling researchers to interpret true relationships between variables. These causal relationships are often represented by causal graphs, which are directed acyclic graphs. With the recent advancements in Large Language Models (LLMs), there is an increasing interest in exploring their capabilities in causal reasoning and their potential use to hypothesize causal graphs. These tasks necessitate the LLMs to encode the causal graph effectively for subsequent downstream tasks. In this paper, we introduce CausalGraph2LLM, a comprehensive benchmark comprising over 700k queries across diverse causal graph settings to evaluate the causal reasoning capabilities of LLMs. We categorize the causal queries into two types: graph-level and node-level queries. We benchmark both open-sourced and propriety models for our study. Our findings reveal that while LLMs show promise in this domain, they are highly sensitive to the encoding used. Even capable models like GPT-4 and Gemini-1.5 exhibit sensitivity to encoding, with deviations of about $60\%$. We further demonstrate this sensitivity for downstream causal intervention tasks. Moreover, we observe that LLMs can often display biases when presented with contextual information about a causal graph, potentially stemming from their parametric memory.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。