Vector Search As Nearest Neighbor Matching: RAG-based Policy Learning in Causal Inference

  • 类型:arxiv
  • 标识:2607.18225
  • 链接:http://arxiv.org/abs/2607.18225v1
  • 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:This work decomposes the regret of the two-step method into candidate-generation regret and within-candidate choice regret, and bound the latter using prediction-error guarantees for nearest-neighbor estimators and transformers.
  • OpenAlex ID:W7169843332
  • OpenAlex DOI:10.48550/arxiv.2607.18225
  • DOI:10.48550/arxiv.2607.18225
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://doi.org/10.48550/arxiv.2607.18225
  • OpenAlex更新:2026-08-24
  • 副分类:llm-infra
  • 待LLM分类:否
  • 标题中文:向量搜索作为最近邻匹配:基于RAG的因果推断策略学习
  • TLDR中文:该工作将两步方法的遗憾分解为候选生成遗憾和候选内选择遗憾,并利用最近邻估计器和Transformer的预测误差保证对后者进行了界。
  • 来源文件
  • /inbox/tom/_candidates/2026-07-21-agent-rag-longcontext-candidates.json
  • [S2 enrich]
  • /inbox/tom/_candidates/2026-07-22-agent-rag-longcontext-candidates.json
  • [OpenAlex backfill]