Efficient Dynamic Clustering-Based Document Compression for Retrieval-Augmented-Generation

ACL ARR 2025 May Submission7907 Authors

Abstract: Retrieval-Augmented Generation (RAG) has emerged as a widely adopted approach for knowledge injection during large language model (LLM) inference in recent years. However, due to their limited ability to exploit fine-grained inter-document relationships, current RAG implementations face challenges in effectively addressing the retrieved noise and redundancy content, which may cause error in the generation results. To address these limitations, we propose an$E$fficient $D$ynamic $C$lustering-based document $C$ompression framework ($EDC^2-RAG$) that utilizes latent inter-document relationships while simultaneously removing irrelevant information and redundant content. We validate our approach, built upon GPT-3.5 and GPT4o-mini, on widely used knowledge-QA and hallucination-detection datasets. Experimental results show that our method achieves consistent performance improvements across various scenarios and experimental settings, demonstrating strong robustness and applicability. Our code and datasets are available at \url{https://anonymous.4open.science/r/EDC-2-RAG-5F54}.

Paper Type: Long

Research Area: Generation

Research Area Keywords: Information Retrieval and Text Mining, Generation, Question Answering

Contribution Types: Approaches low compute settings-efficiency

Languages Studied: English

Submission Number: 7907