CREAM: Continual Retrieval on Dynamic Streaming Corpora with Adaptive Soft Memory

2026-01-06

arXiv.org

AI Summary

This work addresses the challenge of generalizing information retrieval systems in dynamic data streams, where distribution shifts and reliance on fixed labeled queries hinder performance on unseen queries and documents. The paper proposes the first self-supervised continual retrieval framework that operates without ground-truth labels, leveraging a dynamic soft memory module to model the evolving semantics of queries and documents and enabling unsupervised adaptation to both emerging and historical topics. The approach integrates three key components: fine-grained similarity estimation, regularized clustering prototypes, and hierarchical coreset sampling. Evaluated on two standard benchmarks, the method substantially outperforms existing label-free approaches—improving Success@5 by 27.79% and Recall@10 by 44.5%—and achieves performance comparable to, or even surpassing, that of supervised models.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs

Abstract

Information retrieval (IR) in dynamic data streams is a crucial task, as shifts in data distribution degrade the performance of AI-powered IR systems. To mitigate this issue, memory-based continual learning has been widely adopted for IR. However, existing methods rely on a fixed set of queries with ground-truth documents, which limits generalization to unseen data, making them impractical for real-world applications. To enable more effective learning with unseen topics of a new corpus without ground-truth labels, we propose CREAM, a self-supervised framework for memory-based continual retrieval. CREAM captures the evolving semantics of streaming queries and documents into dynamically structured soft memory and leverages it to adapt to both seen and unseen topics in an unsupervised setting. We realize this through three key techniques: fine-grained similarity estimation, regularized cluster prototyping, and stratified coreset sampling. Experiments on two benchmark datasets demonstrate that CREAM exhibits superior adaptability and retrieval accuracy, outperforming the strongest method in a label-free setting by 27.79% in Success@5 and 44.5% in Recall@10 on average, and achieving performance comparable to or even exceeding that of supervised methods.

Problem

• continual retrieval
• dynamic streaming corpora
• information retrieval
• unseen queries
• memory-based continual learning

Innovation

• continual retrieval
• self-supervised learning
• soft memory
• dynamic streaming corpora
• unsupervised adaptation