ERank: Fusing Supervised Fine-Tuning and Reinforcement Learning for Effective and Efficient Text Reranking

Text reranking models are a crucial component in modern systems like Retrieval-Augmented Generation, tasked with selecting the most relevant documents prior to generation. However, current Large Language Models (LLMs) powered rerankers often face a fundamental trade-off. On one hand, Supervised Fine-Tuning based pointwise methods that frame relevance as a binary classification task lack the necessary scoring discrimination, particularly for those built on reasoning LLMs. On the other hand, approaches designed for complex reasoning often employ powerful yet inefficient listwise formulations, rendering them impractical for low latency applications. To resolve this dilemma, we introduce ERANK, a highly Effective and Efficient pointwise reranker built from a reasoning LLM that excels across diverse relevance scenarios. We propose a novel two-stage training pipeline that begins with Supervised Fine-Tuning (SFT). In this stage, we move beyond binary labels and train the model generatively to output fine grained integer scores, which significantly enhances relevance discrimination. The model is then further refined using Reinforcement Learning (RL) with a novel, listwise derived reward. This technique instills global ranking awareness into the efficient pointwise architecture. We evaluate the ERANK reranker on the BRIGHT, FollowIR, TREC DL, and BEIR benchmarks, demonstrating superior effectiveness and robustness compared to existing approaches. On the reasoning-intensive BRIGHT benchmark, our ERANK-4B achieves an nDCG@10 of 38.7, while a larger 32B variant reaches a state of the art nDCG@10 of 40.2. * Corresponding author Positive document: A goldfish will grow to the depth of the water it is kept in … Query: Do goldfish grow? Positive document: … Ramsey's Theorem states that in any coloring of the edges of a sufficiently large complete graph, one will find monochromatic complete subgraphs … Query: In a party, how many guests do you need to have to ensure that either four people all know each other or four people are all complete strangers to one another? Semantic Relevance Reasoning-intensive Relevance Keyword / Semantic matching, example from TREC DL19 Provide essential theorems, example from BRIGHT TheoremQA Theorems