After Retrieval, Before Generation: Enhancing the Trustworthiness of Large Language Models in RAG

2025-05-21 · Xinbang Dai, Huikang Hu, Yuncheng Hua, Jiaqi Li, Yongrui Chen, Rihui Jin, Nan Hu, Guilin Qi

Retrieval-augmented generation (RAG) systems face critical challenges in balancing internal (parametric) and external (retrieved) knowledge, especially when these sources conflict or are unreliable. To analyze these scenarios comprehensively, we construct the Trustworthiness Response Dataset (TRD) with 36,266 questions spanning four RAG settings. We reveal that existing approaches address isolated scenarios-prioritizing one knowledge source, naively merging both, or refusing answers-but lack a unified framework to handle different real-world conditions simultaneously. Therefore, we propose the BRIDGE framework, which dynamically determines a comprehensive response strategy of large language models (LLMs). BRIDGE leverages an adaptive weighting mechanism named soft bias to guide knowledge collection, followed by a Maximum Soft-bias Decision Tree to evaluate knowledge and select optimal response strategies (trust internal/external knowledge, or refuse). Experiments show BRIDGE outperforms baselines by 5-15% in accuracy while maintaining balanced performance across all scenarios. Our work provides an effective solution for LLMs' trustworthy responses in real-world RAG applications.

📄 PDF Abstract BibTeX arXiv:2505.17118

Code (0)

등록된 구현이 없습니다.

Tasks

RAGRetrieval-augmented Generation

Methods 이 논문이 사용한 방법론

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

BERT BERT, or Bidirectional Encoder Representations from Transformers, improves upon standard Transformers by removing the…

BART BART is a denoising autoencoder for pretraining sequence-to-sequence models. It is trained by (1) corrupting text…

RAG Retriever-Augmented Generation, or RAG, is a type of language generation model that combines pre-trained parametric and non-parametric memory for language generation.…

Rethinking All Evidence: Enhancing Trustworthy Retrieval-Augmented Generation via Conflict-Driven Summarization

2025-07-02 · Juan Chen, Baolong Bi, Wei Zhang, Jingyan Sui 외 arxiv

Retrieval-Augmented Generation (RAG) enhances large language models (LLMs) by integrating their parametric knowledge with external retrieved content. However, knowledge conflicts caused by internal inconsistencies or noi…

TrustRAG: An Information Assistant with Retrieval Augmented Generation

2025-02-19 · Yixing Fan, Qiang Yan, Wenshan Wang, Jiafeng Guo 외

\Ac{RAG} has emerged as a crucial technique for enhancing large models with real-time and domain-specific knowledge. While numerous improvements and open-source tools have been proposed to refine the \ac{RAG} framework f…

Answer Generation Chunking Question Answering RAG+2

Trustworthiness in Retrieval-Augmented Generation Systems: A Survey

2024-09-16 · Yujia Zhou, Yan Liu, Xiaoxi Li, Jiajie Jin 외

Retrieval-Augmented Generation (RAG) has quickly grown into a pivotal paradigm in the development of Large Language Models (LLMs). While much of the current research in this field focuses on performance optimization, par…

Fairness Hallucination RAG Retrieval+2

Inference-Time Safety For Code LLMs Via Retrieval-Augmented Revision

2026-03-02 · Manisha Mukherjee, Vincent J. Hellendoorn arxiv

Large Language Models (LLMs) are increasingly deployed for code generation in high-stakes software development, yet their limited transparency in security reasoning and brittleness to evolving vulnerability patterns rais…

Code Generation

LegalCiteTrust: Benchmarking Citation Trustworthiness in Chinese Long-Form Legal Research Reports

2026-07-23 · Yunhan Li, Mingjie Xie, Zeyang Shi, Gengshen Wu 외 arxiv

Long-form legal research reports increasingly rely on LLMs and agentic research systems, but their reliability depends not only on answering the task, but also on whether cited legal authorities are trustworthy. A citati…