The paper introduces a task of logical session complex query answering (LS-CQA) to understand complex user intentions across multiple sessions. These intentions are crucial for enhancing product recommendations, navigation suggestions, and query reformulations. The authors propose a new model, the Logical Session Graph Transformer (LSGT), which captures interactions among items across different sessions and their logical connections using a transformer structure. The LSGT model is evaluated on three datasets and demonstrates state-of-the-art results.
Publication date: 21 Dec 2023
Project Page: https://arxiv.org/abs/2312.13866
Paper: https://arxiv.org/pdf/2312.13866
