通过LLM重排序与证据约束生成，提升多轮RAG的准确性和可信度。
5ting at SemEval-2026 Task 8: Strong End-to-End Multi-Turn RAG via LLM-Based Reranking and Faithfulness Control
- 用双查询融合检索+大模型重排序，增强上下文相关性。
- 端到端系统在任务C中得分0.5597，幻觉率控制在76.92%。
- 适合关注多轮对话生成可靠性与信息溯源的研究者。
我们提出5ting系统，用于SemEval2026任务8（MTRAGEval），评估多轮检索增强生成（RAG）系统。多轮RAG面临上下文漂移、信息不足和幻觉风险。本系统结合BGE-M3稠密检索与FAISS索引，采用双查询融合检索，经由大模型重排序，并通过角色分离生成机制严格依赖检索证据。在任务A中，检索器nDCG@5达0.4719；端到端系统在任务C中获得0.5597的调和平均分，幻觉率（RL_F）为0.7692。
原文摘要 · Abstract (English)
We introduce 5ting, our system for the SemEval2026 Task 8 (MTRAGEval), which evaluates multi-turn Retrieval Augmented Generation (RAG) systems. Multi turn RAG involves context drift, under specification, and hallucination risk. Our system combines BGE-M3 dense retrieval with FAISS indexing, dual-query merged retrieval, and LLM based reranking, followed by role separated generation constrained to retrieved evidence. The retriever achieved nDCG@5 = 0.4719 in Task A, while the end to end system ranked in Task C with a harmonic score of 0.5597 and RL_F = 0.7692.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。