Abstract
Unsupervised dense retrievers offer scalability by learning semantic similarity from unlabeled documents via contrastive learning, but they struggle to capture temporal relevance, retrieving semantically related but temporally misaligned documents—an important aspect when a document collection spans multiple time periods (e.g., retrieving documents from 2018-2025 for "Who is the president in 2019?" introduces temporal ambiguity). Existing methods rely on supervised training with explicit timestamps, which are not always feasible.
We propose TPOUR (Temporal Preference Optimization for Unsupervised Retriever), which uses our novel training method Temporal Retrieval Preference Optimization (TRPO). TRPO reinterprets preference learning in the temporal dimension, guiding the retriever to favor temporally aligned documents. TPOUR further generalizes to unseen time periods via interpolation in a learned time embedding, enabling continuous temporal alignment.
Experiments on temporal information retrieval (T-IR) show that TPOUR outperforms both unsupervised and supervised baselines. Compared to Qwen-Embedding-8B, despite being about 72.7x smaller, TPOUR Contriever improves average nDCG@5 by +4.04 (+12.15%) on explicit and +4.98 (+15.21%) on implicit queries. Our code is available at GitHub.
Blogger's Review: TPOUR introduces a novel approach with temporal preference optimization, effectively addressing the issue of temporal relevance in unsupervised retrieval. Its impressive performance in temporal information retrieval highlights its potential in handling multi-period documents, making it a topic worth watching and further exploration.