1
Answer Presence Drives RAG Rewriting Gains
打破刻板印象：RAG重写提升效果并非来自证据质量改善，而是答案存在性起主导作用
arXiv:2606.05633v1 Announce Type: new Abstract: Retrieval-augmented QA pipelines often route retrieved passages through an LLM \emph{rewriter} before …
2
CoHyDE: Iterative Co-Training of LLM Rewriter & Dense Encoder for Tool Retrieval
用协同训练改写器和密集编码器，跨越自然语言与API技术词汇的鸿沟，大幅提升LLM工具检索准确率。
arXiv:2605.29271v1 Announce Type: new Abstract: Tool retrieval over large API catalogs is a core bottleneck for LLM agents: user queries arrive in col…