Google Research Adds Agentic RAG to Gemini Enterprise Agent Platform with a Sufficient Context Agent for multi-hop queries
English summary
Google Research has introduced a new Agentic RAG framework integrated into the Gemini Enterprise Agent Platform. The framework features a Sufficient Context Agent that iteratively searches until it gathers complete context before generating a response. This multi-agent architecture breaks down complex queries into subtasks, improving accuracy by up to 34% on factuality datasets compared to standard RAG. Tested on the FramesQA benchmark, the system achieved 90.1% accuracy in cross-corpus retrieval while maintaining low latency. The feature, called Cross-Corpus Retrieval, is now in public preview.
Chinese summary
谷歌研究团队推出了一款集成在Gemini企业代理平台中的新型Agentic RAG框架。该框架包含一个“充分上下文代理”，能够反复迭代搜索，直到收集到完整的上下文后再生成响应。这种多代理架构将复杂查询分解为子任务，与标准RAG相比，在事实性数据集上准确率最高提升34%。在FramesQA基准测试中，该系统在跨语料检索中达到90.1%的准确率，同时保持低延迟。该功能名为“跨语料检索”，现已进入公开预览阶段。
Key points
Google introduces Agentic RAG with a Sufficient Context Agent that re-searches until context is complete.
谷歌推出了带有充分上下文代理的Agentic RAG，该代理会反复搜索直到上下文完整。
The multi-agent architecture includes Orchestrator, Planner, Query Rewriter, Search Fanout, Sufficient Context, and Synthesis agents.
多代理架构包括协调器、规划器、查询重写器、搜索扇出、充分上下文和合成代理。
The framework achieves up to 34% higher factuality accuracy over standard RAG.
该框架相比标准RAG，事实性准确率最高提升34%。
Cross-corpus retrieval on FramesQA achieved 90.1% accuracy while selecting from four corpora.
在FramesQA上的跨语料检索实现了90.1%的准确率，同时从四个语料库中选择。
The feature is available as Cross-Corpus Retrieval in Gemini Enterprise Agent Platform in public preview.
该功能以跨语料检索的形式在Gemini企业代理平台上公开预览。