<ul data-eligibleForWebStory="true"><li>Session search typically focuses on sequential modeling for deep semantic understanding, neglecting graph structures in interactions.</li><li>The proposed Symbolic Graph Ranker (SGR) integrates text-based and graph-based approaches using Large Language Models (LLMs).</li><li>SGR converts session graphs into text using symbolic grammar rules, allowing seamless integration of session history, interactions, and task instructions for LLMs.</li><li>The objective is to enhance LLMs' ability to capture graph structures within a textual format.</li><li>Self-supervised symbolic learning tasks like link prediction and node content generation aid LLMs in capturing topological information.</li><li>Experiment results on AOL and Tiangong-ST datasets show the effectiveness of SGR.</li><li>SGR offers a methodology that enhances LLMs in capturing graph structures, bridging traditional search strategies with modern LLMs.</li></ul>

Unify Graph Learning with Text: Unleashing LLM Potentials for Session Search

Discover more