用类人类推理方式优化图数据转文本，提升大模型长距离理解能力
End-to-End Graph Flattening Method for Large Language Models
- 基于有向无环图路径设计端到端提示，模仿人类推理顺序
- 在长距离场景下显著提升大模型推理准确率，短距离表现亦优秀
- 适合需要复杂关系推理的图数据应用，如知识图谱、社交网络分析
近年来，大型语言模型（LLMs）为图数据的通用处理提供了新思路。将图转化为自然语言（即图扁平化）具有良好的泛化性和可解释性，但文本格式组织不佳，导致长距离场景下的理解性能较差。受人类认知推理习惯启发，我们提出一种新型图扁平化方法——端到端有向无环图路径提示（End-to-End DAG-Path prompting, EEDP），以适配大模型。在真实世界数据集上的实验表明，EEDP 在长距离场景中显著提升了大模型的推理性能，同时保持了短距离场景下的优异表现，展现出对距离变化的良好鲁棒性。
原文摘要 · Abstract (English)
In recent years, the breakthrough of Large Language Models (LLMs) offers new ideas for achieving universal methods on graph data. The common practice of converting graphs into natural language for LLMs, which refers to graph flattening, exhibits good generalizability and interpretability. However, the poor organization of the textual format results in poor performance in long-distance scenario understanding. Inspired by human cognitive reasoning habits, we propose a novel method for graph flattening to fit LLMs, termed as End-to-End DAG-Path prompting (EEDP). Experiments on real-world datasets show that EEDP enhances the reasoning performance of LLMs in long-distance scenarios while maintaining excellent performance in short-distance scenarios, demonstrating good robustness in the face of distance variations.
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。