RAG 研究聚焦成本、意图和分块以改进 AI 检索

报道来源 [4]

1. arXiv cs.AI TIER_1 English(EN) · Sanjay Mishra · 2026-06-03 04:00

   RAG中的成本感知查询路由：检索深度权衡的实证分析

   arXiv:2606.02581v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) faces a fundamental three-way tension: deeper retrieval improves factual grounding but inflates token costs and end-to-end latency. Static retrieval configurations cannot resolve this tension a…

2. arXiv cs.CL TIER_1 English(EN) · Fachrina Dewi Puspitasari, Chaoning Zhang, Jiaquan Zhang, Zhicheng Wang, Hafiz Shakeel Ahmad Awan, Rizwan Qureshi, Jewon Lee, Tae-Ho Kim, Yang Yang · 2026-06-02 04:00

   具有意图感知检索和语义保留分块的高效 RAG

   arXiv:2606.01240v1 Announce Type: new Abstract: The demand for powerful instruction following and reasoning capability of large language models (LLMs) has promoted rapid development of retrieval-augmented generation (RAG). The RAG system assists LLM generation by retrieving chunk…

3. dev.to — LLM tag TIER_1 English(EN) · Vipul · 2026-06-01 15:53

   RAG中的分块为何重要：更好检索的隐藏关键

   <p>When people discuss Retrieval-Augmented Generation (RAG), they often focus on embeddings, vector databases, or LLMs. However one of the most critical factors affecting RAG performance is chunking.</p> <p>A well-designed chunking strategy can significantly improve retrieval acc…

4. dev.to — LLM tag TIER_1 English(EN) · kartikey rajvaidya · 2026-06-01 11:10

   免费上下文分块标题：面向混合检索的感知标题分块

   <p>In September 2024, Anthropic published <em>Contextual Retrieval</em>. The trick: generate a one-sentence context per chunk with an LLM and prepend it to the chunk before embedding. On their hybrid vector + BM25 setup, the top-20 retrieval failure rate drops from 5.7% to 2.9% (…