CQC-RAG: Robust Retrieval-Augmented Generation via Cross-Query Consistency

  • 类型:arxiv
  • 标识:2606.13438
  • 链接:http://arxiv.org/abs/2606.13438v1
  • 主分类:rag
  • 形态:method
  • 被引:0
  • 被引来源:Semantic Scholar + OpenAlex
  • S2被引:0
  • OpenAlex被引:0
  • 影响力被引:0
  • TLDR:CQC-RAG, a framework that co-designs query-level diversity injection with cross-query consistency evaluation and enables self-evaluation without external supervision, is introduced, validating the effectiveness of cross-query consistency for filtering noise-induced hallucinations.
  • OpenAlex ID:W7164582822
  • OpenAlex DOI:10.48550/arxiv.2606.13438
  • DOI:10.48550/arxiv.2606.13438
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://doi.org/10.48550/arxiv.2606.13438
  • OpenAlex更新:2026-07-19
  • 待LLM分类:否
  • 副分类:evaluation
  • 标题中文:CQC-RAG:通过跨查询一致性实现鲁棒的检索增强生成
  • TLDR中文:CQC-RAG 框架被提出,它协同设计查询级多样性注入与跨查询一致性评估,无需外部监督即可实现自我评估,验证了跨查询一致性在过滤噪声引发幻觉方面的有效性。
  • 来源文件
  • /inbox/tom/_candidates/2026-06-17-rag-retrieval-reranking-candidates.json
  • [S2 enrich]
  • [OpenAlex backfill]