SeCon-RAG: A Two-Stage Semantic Filtering and Conflict-Free Framework for Trustworthy RAG

Xinfeng Li (Nanyang Technological University) · Yang Liu (CUHK) · Lijun Zhang (Nanjing University (NJU)) · Xiaojun Jia (Nanyang Technological University) · Simeng Qin (Northeast University at Qinhuangdao Campus) · Lijia Yu (Institute of Software, Chinese Academy of Sciences) · Xiaonan Si (cas) · Meilin Zhu (Institute of Software, Chinese Academy of Sciences) · Shuaitong Liu (Southwest University) · Ranjie Duan (Swinburne University of Technology)

candidate answersclean retrieval databaseconflict-aware filteringconflict-free frameworkcontamination attackscorpus poisoningentity-intent-relation extractorgeneration robustnessknowledge retrievallatent objectivesoutput trustworthinessretrieval-augmented generation semantic consistencysemantic filteringsemantic relevance

Retrieval-augmented generation (RAG) systems enhance large language models (LLMs) with external knowledge but are vulnerable to corpus poisoning and contamination attacks, which can compromise output integrity. Existing defenses often apply aggressive filtering, leading to unnecessary loss of valuable information and reduced reliability in generation. To address this problem, we propose a two-stage semantic filtering and conflict-free framework for trustworthy RAG. In the first stage, we perform a joint filter with semantic and cluster-based filtering which is guided by the Entity-intent-relation extractor (EIRE). EIRE extracts entities, latent objectives, and entity relations from both the user query and filtered documents, scores their semantic relevance, and selectively adds valuable documents into the clean retrieval database. In the second stage, we proposed an EIRE-guided conflict-aware filtering module, which analyzes semantic consistency between the query, candidate answers, and retrieved knowledge before final answer generation, filtering out internal and external contradictions that could mislead the model. Through this two-stage process, SeCon-RAG effectively preserves useful knowledge while mitigating conflict contamination, achieving significant improvements in both generation robustness and output trustworthiness. Extensive experiments across various LLMs and datasets demonstrate that the proposed SeCon-RAG markedly outperforms state-of-the-art defense methods.