[论文解读] Causality for Large Language Models

Anpeng Wu, Kun Kuang+8 more|arXiv (Cornell University)|Oct 20, 2024

Topic Modeling被引用 5

一句话总结

对在预训练、微调、对齐、推理和评估中将因果关系整合进大语言模型的全面综述，以克服对虚假相关和偏见的依赖。

ABSTRACT

Recent breakthroughs in artificial intelligence have driven a paradigm shift, where large language models (LLMs) with billions or trillions of parameters are trained on vast datasets, achieving unprecedented success across a series of language tasks. However, despite these successes, LLMs still rely on probabilistic modeling, which often captures spurious correlations rooted in linguistic patterns and social stereotypes, rather than the true causal relationships between entities and events. This limitation renders LLMs vulnerable to issues such as demographic biases, social stereotypes, and LLM hallucinations. These challenges highlight the urgent need to integrate causality into LLMs, moving beyond correlation-driven paradigms to build more reliable and ethically aligned AI systems. While many existing surveys and studies focus on utilizing prompt engineering to activate LLMs for causal knowledge or developing benchmarks to assess their causal reasoning abilities, most of these efforts rely on human intervention to activate pre-trained models. How to embed causality into the training process of LLMs and build more general and intelligent models remains unexplored. Recent research highlights that LLMs function as causal parrots, capable of reciting causal knowledge without truly understanding or applying it. These prompt-based methods are still limited to human interventional improvements. This survey aims to address this gap by exploring how causality can enhance LLMs at every stage of their lifecycle-from token embedding learning and foundation model training to fine-tuning, alignment, inference, and evaluation-paving the way for more interpretable, reliable, and causally-informed models. Additionally, we further outline six promising future directions to advance LLM development, enhance their causal reasoning capabilities, and address the current limitations these models face.

研究动机与目标

• 激励将因果推理整合到大语言模型中，以克服基于相关性的局限性和偏见。
• 提出一个生命周期框架，展示如何在预训练、微调、对齐、推理和评估阶段将因果性纳入其中。
• 对基于因果性的技术进行分类，并将它们映射到大语言模型开发的各个阶段。
• 突出挑战并为因果信息化的大语言模型提出未来方向。

提出的方法

• 回顾在五个大语言模型阶段：预训练、微调、对齐、推理和评估中现有的因果启发技术。
• 提出三种因果预训练方法：去偏置的标记嵌入、反事实训练语料以及因果基础模型框架。
• 概述对微调、对齐和推理的因果增强过程，包括因果发现、因果效应和反事实推理。

实验结果

研究问题

• RQ1 如何将因果性嵌入到大语言模型的训练和生命周期中，以超越基于相关性的学习？
• RQ2 在各阶段（预训练、微调、对齐、推理、评估）中，哪些技术最能促进大语言模型的因果推理？
• RQ3 构建因果知情的大语言模型的主要挑战和未来方向有哪些？

主要发现

• 跨阶段的因果性技术可以解决大语言模型中的偏见、可靠性和可解释性问题。
• 当前的因果提示依赖于人类干预和提示，而非模型内在理解；将因果性嵌入到训练中是必不可少的。
• 提出六个未来方向，以推动因果推理并克服大语言模型的局限性。