研究人员开发了RASER，一个旨在通过减少昂贵的检索调用次数来优化多跳问答的新系统。RASER根据来自初始一次性RAG的六个特征，仅在必要时选择性地升级到更复杂的检索方法。这种方法显著降低了token成本，使用的token比总是修剪的方法少41-49%，同时在各种LLM和基准测试中保持了具有竞争力的准确性。 AI
影响 降低了复杂问答任务的计算成本，提高了LLM应用程序的效率。
排序理由 该集群包含一篇详细介绍问答新方法的论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2606.02488v1 Announce Type: new Abstract: Multi-hop question-answering systems often use expensive retrieval on every question. They may decompose the question, run several retrieval rounds, or search through bridge entities before answering. All of these strategies rely on…
Multi-hop question-answering systems often use expensive retrieval on every question. They may decompose the question, run several retrieval rounds, or search through bridge entities before answering. All of these strategies rely on repeated LLM calls to rewrite or decompose the …