Embedding-Based Context-Aware Reranker
On the map: Retrieval-augmented generation → Generation and context integration
Abstract (arXiv)
Retrieval-Augmented Generation (RAG) systems rely on retrieving relevant evidence from a corpus to support downstream generation. The common practice of splitting a long document into multiple shorter passages enables finer-grained and targeted information retrieval. However, it also introduces challenges when a correct retrieval would require inference across passages, such as resolving coreference, disambiguating entities, and aggregating evidence scattered across multiple sources. Many state-of-the-art (SOTA) reranking methods, despite utilizing powerful large pretrained language models with potentially high inference costs, still neglect the aforementioned challenges. Therefore, we propose Embedding-Based Context-Aware Reranker (EBCAR), a lightweight reranking framework operating directly on embeddings of retrieved passages with enhanced cross-passage understandings through the structural information of the passages and a hybrid attention mechanism, which captures both high-level interactions across documents and low-level relationships within each document. We evaluate EBCAR against SOTA rerankers on the ConTEB benchmark, demonstrating its effectiveness for information retrieval requiring cross-passage inference and its advantages in both accuracy and efficiency.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Generation and context integration, one of 15 directions in Retrieval-augmented generation. Builds on 36 of its 43 references that are in the map. First posted on arXiv 15 Oct 2025.
Builds on
36 of 43 references.
Closest ancestors · 2
Same direction, newest first
- 2024Attention in Large Language Models Yields Efficient Zero-Shot Re-Rankers3 Oct
- 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks22 May
Other references · 34
- 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
- 2016MS MARCO: A Human Generated MAchine Reading COmprehension Datasetcited by 1,605
- 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,253
- 2021BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Modelscited by 1,252
- 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,216
- 2024Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inferencecited by 554
- 2020Document Ranking with a Pretrained Sequence-to-Sequence Modelcited by 376
- 2023Is ChatGPT Good at Search? Investigating Large Language Models as Re-Ranking Agentscited by 360
- 2020Overview of the TREC 2019 deep learning trackcited by 286
- 2024mGTE: Generalized Long-Context Text Representation and Reranking Models for Multilingual Text Retrievalcited by 236
- 2019Multi-Stage Document Ranking with BERTcited by 201
- 2021Overview of the TREC 2020 deep learning trackcited by 196
22 more
- 2020Generation-Augmented Retrieval for Open-domain Question Answeringcited by 186
- 2023RankZephyr: Effective and Robust Zero-Shot Listwise Reranking is a Breeze!cited by 155
- 2023Retrieval meets Long Context Large Language Modelscited by 115
- 2022Re2G: Retrieve, Rerank, Generatecited by 102
- 2024Retrieval-Augmented Generation with Knowledge Graphs for Customer Service Question Answeringcited by 96
- 2021The Expando-Mono-Duo Design Pattern for Text Ranking with Pretrained Sequence-to-Sequence Modelscited by 81
- 2024Late Chunking: Contextual Chunk Embeddings Using Long-Context Embedding Modelscited by 73
- 2023RankVicuna: Zero-Shot Listwise Document Reranking with Open-Source Large Language Modelscited by 70
- 2024FIRST: Faster Improved Listwise Reranking with Single Token Decodingcited by 54
- 2024LongEmbed: Extending Embedding Models for Long Context Retrievalcited by 48
- 2024Is Semantic Chunking Worth the Computational Cost?cited by 43
- 2024UniMS-RAG: A Unified Multi-source Retrieval-Augmented Generation for Personalized Dialogue Systemscited by 38
- 2021Simplified Data Wrangling with ir_datasetscited by 27
- 2023Scaling Down, LiTting Up: Efficient Zero-Shot Listwise Reranking with Seq2seq Encoder-Decoder Modelscited by 26
- 2024Retrieval Augmented Fact Verification by Synthesizing Contrastive Argumentscited by 25
- 2024Leveraging Passage Embeddings for Efficient Listwise Reranking with Large Language Modelscited by 20
- 2024Learning Retrieval Augmentation for Personalized Dialogue Generationcited by 18
- 2024Enhancing Q&A Text Retrieval with Ranking Models: Benchmarking, fine-tuning and deploying Rerankers for RAGcited by 16
- 2025RankLLM: A Python Package for Reranking with LLMscited by 16
- 2025Context is Gold to find the Gold Passage: Evaluating and Training Contextual Document Embeddingscited by 13
- 2025Hierarchical Retrieval with Evidence Curation for Open-Domain Financial Question Answering on Standardized Documentscited by 5
- 2024An Early FIRST Reproduction and Improvements to Single-Token Decoding for Fast Listwise Rerankingcited by 4
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Beyond the Reranker: Do RAG Retrieval Enhancements Help Once a Strong Reranker Is Present?
- 2025Enhancing Retrieval-Augmented Generation with Two-Stage Retrieval: FlashRank Reranking and Query Expansion
- 2025The Evolution of Reranking Models in Information Retrieval: From Heuristic Methods to Large Language Models
- 2026Lychee-Reranker: Fast but Not Late Interaction for Compressed Document Reranking
- 2024Don't Forget to Connect! Improving RAG with Graph-based Reranking
- 2026EmbeddingRWKV: State-Centric Retrieval with Reusable States
- 2025From Ranking to Selection: A Simple but Efficient Dynamic Passage Selector for Retrieval Augmented Generation
- 2024JudgeRank: Leveraging Large Language Models for Reasoning-Intensive Reranking