CiteCheck: Towards Accurate Citation Faithfulness Detection

TL;DR

CiteCheck introduces the first large-scale Chinese dataset for citation faithfulness detection, enabling more accurate and cost-effective training of models, and highlights the challenge of high difficulty in test samples for current systems.

Contribution

This paper presents a novel large-scale Chinese dataset for citation faithfulness detection, created via a cost-effective two-stage manual annotation process, and demonstrates its effectiveness for training models.

Findings

Test samples are highly challenging for state-of-the-art LLMs.

Augmenting training data with LLM-generated negatives improves model performance.

The dataset facilitates research in Chinese citation faithfulness detection.

Abstract

Citation faithfulness detection is critical for enhancing retrieval-augmented generation (RAG) systems, yet large-scale Chinese datasets for this task are scarce. Existing methods face prohibitive costs due to the need for manually annotated negative samples. To address this, we introduce the first large-scale Chinese dataset CiteCheck for citation faithfulness detection, constructed via a cost-effective approach using two-stage manual annotation. This method balances positive and negative samples while significantly reducing annotation expenses. CiteCheck comprises training and test splits. Experiments demonstrate that: (1) the test samples are highly challenging, with even state-of-the-art LLMs failing to achieve high accuracy; and (2) training data augmented with LLM-generated negative samples enables smaller models to attain strong performance using parameter-efficient fine-tuning. CiteCheck provides a robust foundation for advancing citation faithfulness detection in Chinese RAG systems. The dataset is publicly available to facilitate research.