面试题库

全部 Django Elasticsearch FastAPI Flask Function Call JVM Java IO Java 后端 Java 基础 Java 并发 Java 新特性 Java 集合 JavaWeb Kafka LLM LangChain LangGraph MongoDB MyBatis MySQL Netty Nginx Python RAG RabbitMQ Redis RocketMQ Spring Spring AI SpringBoot SpringCloud SpringMVC ZooKeeper 分布式后端技术大厂面经大数据微服务提示词工程操作系统数据库数据结构智能体海康威视2026秋季社招专项消息队列监控运维算法系统设计缓存计算机基础计算机网络设计模式面试题高可用高并发

RAG 中等

Is RAG still relevant in the era of long context LLMs?

作答 →

RAG 中等

Explain why RAG is required when LLMs are already powerful

作答 →

RAG 中等

In what situations would you prioritize Context Precision over Context Recall in a RAG retriever, and how would this impact the generator’s performance?

作答 →

RAG 中等

If your retriever achieves high context precision but low context recall, what types of user queries would likely suffer most?

作答 →

RAG 中等

In a RAG pipeline, how might context recall impact the completeness of generated answers? Describe a scenario illustrating this relationship

作答 →

RAG 中等

What happens with a weak retriever in Retrieval-Augmented Generation (RAG) systems?

作答 →

RAG 中等

What are the possible reasons for the poor performance of a RAG retriever?

作答 →

RAG 中等

How does the chunking strategy differ when dealing with structured documents (like PDFs with tables and figures) versus plain text documents?

作答 →