RankZephyr: Effective and Robust Zero-Shot Listwise Reranking is a Breeze!

By Ronak Pradeep, Sahel Sharifymoghaddam, Jimmy J. Lin. 2023

Subfields: Information Retrieval & Search, Large Language Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Information Retrieval & Search

1. Wei-Chao Zhou
2. Wen-Chao Li
3. Jiaxin Zhang
4. Anu Singh
5. Hilaf Hasson

Citations in the corpus

• CoRNStack: High-Quality Contrastive Data for Better Code Retrieval and Reranking (2024, unknown): “Recently, listwise reranking approaches (Pradeep et al., 2023; Reddy et al., 2024) have gained popularity for their ability to score multiple passages simultaneously, as opposed to pointwise Zhuang et al. (2023a;c) or pairwise Qin et al. (2023) reranking, where scoring is performed in isolation.”
• What Drives Cross-lingual Ranking? Retrieval Approaches with Multilingual Language Models (2025, unknown): “Open-source variants such as RankVicuna and RankZephyr [55, 56], and distillation-based methods like Rank-without-GPT [57], demonstrate competitive performance on English benchmarks.”
• SweRank: Software Issue Localization with Code Ranking (2025, unknown): “S WE R ANK LLM adopts a listwise ranking approach (Pradeep et al., 2023b; Reddy et al., 2024), which generally offers better performance than pointwise methods by considering the relative relevance of candidates.”
• HyQE: Ranking Contexts with Hypothetical Query Embeddings (2024, methodology): “For instance, LLM-based re-rankers (Sun et al., 2023; Pradeep et al., 2023) can determine whether a context addresses a query better than others.”