Ground Every Sentence: Improving Retrieval-Augmented LLMs with Interleaved Reference-Claim Generation

Sirui Xia, Xintao Wang, Jiaqing Liang, Yifei Zhang, Weikang Zhou +3 more
7/1/2024
cs.CL

No official code found. Best implementation: pdxthree/ReClaim (3 stars, Apache-2.0, last commit 2025-05-26).

Abstract

Retrieval-Augmented Generation (RAG) has been widely adopted to enhance Large Language Models (LLMs) in knowledge-intensive tasks. To enhance credibility and verifiability in RAG systems, Attributed Text Generation (ATG) is proposed, which provides citations to retrieval knowledge in LLM-generated responses. Prior methods mainly adopt coarse-grained attributions, with passage-level or paragraph-level references or citations, which fall short in verifiability. This paper proposes ReClaim (Refer & Claim), a fine-grained ATG method that alternates the generation of references and answers step by step. Different from previous coarse-grained attribution, ReClaim provides sentence-level citations in long-form question-answering tasks. With extensive experiments, we verify the effectiveness of ReClaim in extensive settings, achieving a citation accuracy rate of 90%.

View on arXivView PDF

Code Implementations(1)

pdxthree/ReClaimCommunity57%
Apache-2.0

Here is the open-source code repository for the paper "Ground Every Sentence: Improving Retrieval-Augmented LLMs with Interleaved Reference-Claim Generation."

30PythonFeb 7, 20251 years agoApache-2.0

Similar Papers

Concise and Sufficient Sub-Sentence Citations for Retrieval-Augmented Generation

89% match2025

On the Capacity of Citation Generation by Large Language Models

88% match2024

LLM-Ref: Enhancing Reference Handling in Technical Writing with Large Language Models

86% match2024

Attribute First, then Generate: Locally-attributable Grounded Text Generation

86% match1 repo302024

RAG-E: Quantifying Retriever-Generator Alignment and Failure Modes

84% match2026

CiteFix: Enhancing RAG Accuracy Through Post-Processing Citation Correction

84% match2025

Cite this paper

@article{xia2024ground,
  title  = {Ground Every Sentence: Improving Retrieval-Augmented LLMs with Interleaved Reference-Claim Generation},
  author = {Sirui Xia and Xintao Wang and Jiaqing Liang and Yifei Zhang and Weikang Zhou and Jiaji Deng and Fei Yu and Yanghua Xiao},
  year   = {2024},
  eprint = {2407.01796},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2407.01796v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2407.01796.svg)](https://codeofpaper.com/paper/2407.01796)

Discussion