Efficient Retrieval of Temporal Event Sequences from Textual Descriptions

Abstract

Retrieving temporal event sequences from textual descriptions is essential for applications such as analyzing e-commerce behavior, monitoring social media activities, and tracking criminal incidents. In this paper, we introduce TPP-LLM-Embedding, a unified model for efficiently embedding and retrieving event sequences based on natural language descriptions. Built on the TPP-LLM framework, which integrates large language models with temporal point processes, our model encodes both event types and times, generating a sequence-level representation through pooling. Textual descriptions are embedded using the same architecture, ensuring a shared embedding space for both sequences and descriptions. We optimize a contrastive loss based on similarity between these embeddings, bringing matching pairs closer and separating non-matching ones. TPP-LLM-Embedding enables efficient retrieval and demonstrates superior performance compared to baseline models across diverse datasets.

Recommended citation: Liu, Zefang and Quan, Yinzhu. "Efficient Retrieval of Temporal Event Sequences from Textual Descriptions." arXiv preprint arXiv:2410.14043 (2024).
[Download Paper] [Download Code] [Download Data]