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
LLM 中等

What are the challenges in performing distributed inference across multiple GPUs?

作答 →
LLM 中等

What is speculative decoding, and when would you use it?

作答 →
LLM 中等

Explain the pros and cons of large and small context windows in LLM inference

作答 →
LLM 中等

What is the role of the context window during LLM inference?

作答 →
LLM 中等

Compare deterministic and stochastic decoding methods in LLMs

作答 →
LLM 中等

When might prompt engineering be preferred over task-specific fine-tuning?

作答 →
LLM 中等

Explain how parameter efficient fine-tuning addresses the limitations of full fine-tuning

作答 →
LLM 中等

What are the strengths and limitations of full fine-tuning?

作答 →
LLM 中等

After tokenization, how are tokens converted into embeddings in the Transformer model?

作答 →
LLM 中等

How do you handle the large memory requirements of KV cache in LLM inference?

作答 →
LLM 中等

How does quantization affect inference speed and memory requirements?

作答 →
LLM 中等

What is the role of self-attention in the Transformer model, and why is it called “self-attention”?

作答 →