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

Why is the cross-encoder typically used as the re-ranker rather than the bi-encoder?

作答 →
RAG 中等

Compare general re-rankers and instruction-following re-rankers in RAG

作答 →
RAG 中等

What are the key metrics for evaluating retrieval quality in RAG?

作答 →
RAG 中等

What are some common challenges in RAG retrieval?

作答 →
RAG 中等

What are the common retrieval approaches used in RAG systems?

作答 →
RAG 中等

Explain how the HyDE query transformation technique works

作答 →
RAG 中等

What are the pros and cons of query transformation techniques?

作答 →
RAG 中等

What are the different query transformation techniques that enhance user queries in RAG?

作答 →
RAG 中等

Describe the vector pre-computation and storage strategy in a bi-encoder + cross-encoder pipeline. Why can't cross-encoders pre-compute text representations like bi-encoders can?

作答 →
RAG 中等

If your RAG system serves both simple factual queries and complex analytical questions, how would you decide when to bypass the re-ranker for efficiency while maintaining quality?

作答 →
RAG 中等

Describe a scenario where a BM25 retrieval might return relevant chunks but in poor ranking order. How would a neural re-ranker specifically address this limitation?

作答 →
RAG 中等

What is the significance of Context Precision@K in evaluating a RAG retriever, and how does it differ from standard Precision@k in traditional information retrieval?

作答 →