Not Just What, But When: Integrating Irregular Intervals to LLM for Sequential Recommendation

TL;DR

This paper introduces IntervalLLM, a framework that incorporates dynamic interval information into large language models for sequential recommendation, improving performance especially in cold-start scenarios by considering time intervals between user actions.

Contribution

The work proposes a novel interval-infused attention mechanism in LLMs for sequential recommendation, emphasizing the importance of interval information in both warm and cold scenarios.

Findings

Achieves 4.4% average improvement over baselines.

Best performance in warm and cold scenarios across benchmarks.

Interval perspective reveals significant cold-start challenges.

Abstract

Time intervals between purchasing items are a crucial factor in sequential recommendation tasks, whereas existing approaches focus on item sequences and often overlook by assuming the intervals between items are static. However, dynamic intervals serve as a dimension that describes user profiling on not only the history within a user but also different users with the same item history. In this work, we propose IntervalLLM, a novel framework that integrates interval information into LLM and incorporates the novel interval-infused attention to jointly consider information of items and intervals. Furthermore, unlike prior studies that address the cold-start scenario only from the perspectives of users and items, we introduce a new viewpoint: the interval perspective to serve as an additional metric for evaluating recommendation methods on the warm and cold scenarios. Extensive experiments on 3 benchmarks with both traditional- and LLM-based baselines demonstrate that our IntervalLLM achieves not only 4.4% improvements in average but also the best-performing warm and cold scenarios across all users, items, and the proposed interval perspectives. In addition, we observe that the cold scenario from the interval perspective experiences the most significant performance drop among all recommendation methods. This finding underscores the necessity of further research on interval-based cold challenges and our integration of interval information in the realm of sequential recommendation tasks. Our code is available here: https://github.com/sony/ds-research-code/tree/master/recsys25-IntervalLLM.