Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2409.03258 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916524517752832 |
|---|---|
| author | Cao, Yukun Han, Shuo Gao, Zengyi Ding, Zezhong Xie, Xike Zhou, S. Kevin |
| author_facet | Cao, Yukun Han, Shuo Gao, Zengyi Ding, Zezhong Xie, Xike Zhou, S. Kevin |
| contents | Although Large Language Models (LLMs) have demonstrated potential in processing graphs, they struggle with comprehending graphical structure information through prompts of graph description sequences, especially as the graph size increases. We attribute this challenge to the uneven memory performance of LLMs across different positions in graph description sequences, known as ''positional biases''. To address this, we propose GraphInsight, a novel framework aimed at improving LLMs' comprehension of both macro- and micro-level graphical information. GraphInsight is grounded in two key strategies: 1) placing critical graphical information in positions where LLMs exhibit stronger memory performance, and 2) investigating a lightweight external knowledge base for regions with weaker memory performance, inspired by retrieval-augmented generation (RAG). Moreover, GraphInsight explores integrating these two strategies into LLM agent processes for composite graph tasks that require multi-step reasoning. Extensive empirical studies on benchmarks with a wide range of evaluation tasks show that GraphInsight significantly outperforms all other graph description methods (e.g., prompting techniques and reordering strategies) in understanding graph structures of varying sizes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_03258 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GraphInsight: Unlocking Insights in Large Language Models for Graph Structure Understanding Cao, Yukun Han, Shuo Gao, Zengyi Ding, Zezhong Xie, Xike Zhou, S. Kevin Computation and Language Although Large Language Models (LLMs) have demonstrated potential in processing graphs, they struggle with comprehending graphical structure information through prompts of graph description sequences, especially as the graph size increases. We attribute this challenge to the uneven memory performance of LLMs across different positions in graph description sequences, known as ''positional biases''. To address this, we propose GraphInsight, a novel framework aimed at improving LLMs' comprehension of both macro- and micro-level graphical information. GraphInsight is grounded in two key strategies: 1) placing critical graphical information in positions where LLMs exhibit stronger memory performance, and 2) investigating a lightweight external knowledge base for regions with weaker memory performance, inspired by retrieval-augmented generation (RAG). Moreover, GraphInsight explores integrating these two strategies into LLM agent processes for composite graph tasks that require multi-step reasoning. Extensive empirical studies on benchmarks with a wide range of evaluation tasks show that GraphInsight significantly outperforms all other graph description methods (e.g., prompting techniques and reordering strategies) in understanding graph structures of varying sizes. |
| title | GraphInsight: Unlocking Insights in Large Language Models for Graph Structure Understanding |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2409.03258 |