arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.04560cs.LG
收藏

像图一样表达:面向大语言模型的图编码方法

Talk like a Graph: Encoding Graphs for Large Language Models

Google Research(谷歌研究院)

Bahare Fatemi, Jonathan Halcrow, Bryan Perozzi

AI总结:

本研究首次全面研究面向大语言模型的图文本编码方法,发现编码方式、任务类型、图结构均影响推理性能,据此选择合适编码器可使图推理性能提升4.8%至61.8%。

摘要:

图是在社交网络、推荐系统、计算金融等真实场景中表示与分析复杂关系的有力工具。图推理对于推断复杂系统中实体间的关系、识别隐藏模式与趋势至关重要。尽管自然文本自动推理已取得显著进展,但利用大语言模型(LLMs)开展图推理仍是一个研究不足的问题。本研究首次针对将图结构数据编码为文本以供LLMs处理的问题展开全面研究。我们发现,LLM在图推理任务上的性能在三个基本层面存在差异:(1)图编码方法,(2)图任务本身的性质,(3)值得注意的是,所研究图的自身结构。这些新发现为将图编码为文本的策略提供了宝贵见解。基于这些见解,我们证明选择合适的编码器可使LLM内部图推理任务的性能提升4.8%至61.8%,提升幅度因任务而异。

英文摘要:

Graphs are a powerful tool for representing and analyzing complex relationships in real-world applications such as social networks, recommender systems, and computational finance. Reasoning on graphs is essential for drawing inferences about the relationships between entities in a complex system, and to identify hidden patterns and trends. Despite the remarkable progress in automated reasoning with natural text, reasoning on graphs with large language models (LLMs) remains an understudied problem. In this work, we perform the first comprehensive study of encoding graph-structured data as text for consumption by LLMs. We show that LLM performance on graph reasoning tasks varies on three fundamental levels: (1) the graph encoding method, (2) the nature of the graph task itself, and (3) interestingly, the very structure of the graph considered. These novel results provide valuable insight on strategies for encoding graphs as text. Using these insights we illustrate how the correct choice of encoders can boost performance on graph reasoning tasks inside LLMs by 4.8% to 61.8%, depending on the task.

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。

↑