$ ext{E}^2 ext{Rank}$: Your Text Embedding can Also be an Effective and Efficient Listwise Reranker

2025-10-26

AI Summary

Text embedding models offer efficiency but suffer from limited ranking accuracy, falling short of computationally intensive LLM-based re-rankers. To address this, we propose E²Rank—a novel framework that enables a single embedding model to jointly support retrieval and listwise re-ranking. E²Rank achieves this by continuously fine-tuning text embedding models on listwise ranking objectives, using cosine similarity as the unified scoring function. It further introduces a candidate-aware query augmentation mechanism that simulates pseudo-relevance feedback, explicitly modeling both query–document and document–document interactions. Crucially, E²Rank preserves low latency and high throughput while substantially improving re-ranking quality. Experiments demonstrate that E²Rank achieves state-of-the-art performance on the BEIR re-ranking benchmark, excels in the inference-heavy BRIGHT benchmark, and even enhances performance on the MTEB embedding benchmark—showcasing its dual utility for both retrieval and re-ranking tasks.

Technology Category

• Machine Learning: Learning Preferences or Rankings
• Search and Optimization: Learning to Search
• Computer Vision: Image and Video Retrieval

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
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Text embedding models serve as a fundamental component in real-world search applications. By mapping queries and documents into a shared embedding space, they deliver competitive retrieval performance with high efficiency. However, their ranking fidelity remains limited compared to dedicated rerankers, especially recent LLM-based listwise rerankers, which capture fine-grained query-document and document-document interactions. In this paper, we propose a simple yet effective unified framework $ ext{E}^2 ext{Rank}$, means Efficient Embedding-based Ranking (also means Embedding-to-Rank), which extends a single text embedding model to perform both high-quality retrieval and listwise reranking through continued training under a listwise ranking objective, thereby achieving strong effectiveness with remarkable efficiency. By applying cosine similarity between the query and document embeddings as a unified ranking function, the listwise ranking prompt, which is constructed from the original query and its candidate documents, serves as an enhanced query enriched with signals from the top-K documents, akin to pseudo-relevance feedback (PRF) in traditional retrieval models. This design preserves the efficiency and representational quality of the base embedding model while significantly improving its reranking performance. Empirically, $ extrm{E}^2 ext{Rank}$ achieves state-of-the-art results on the BEIR reranking benchmark and demonstrates competitive performance on the reasoning-intensive BRIGHT benchmark, with very low reranking latency. We also show that the ranking training process improves embedding performance on the MTEB benchmark. Our findings indicate that a single embedding model can effectively unify retrieval and reranking, offering both computational efficiency and competitive ranking accuracy.

Problem

• Extends text embedding models to perform retrieval and reranking
• Improves ranking fidelity while maintaining computational efficiency
• Unifies retrieval and reranking using listwise ranking objective

Innovation

• Extends embedding model for retrieval and reranking
• Uses listwise ranking objective with continued training
• Applies cosine similarity as unified ranking function