Сравните Qdrant, Milvus, Weaviate, Pinecone, pgvector по фичам и характеристикам.
Machine Learning / AI
What LLM evaluation benchmarks do you know (MMLU, BBH, GSM8K, HumanEval)?
What is a video-language model (VideoLLaMA, Video-ChatGPT)?
What are slowly changing dimensions (SCD) and how do they affect ML features?
What is Retrieval-Augmented Generation? What components does it consist of?
What is a decision transformer and how does it translate RL into supervised learning?
What are the bottlenecks when deploying GNN in production?
What is RT-DETR and what is its idea?
What is calibration and why is it important for fairness?
How to obtain a sentence vector from BERT embeddings? Compare CLS, mean pooling, max pooling.
What is Apple Neural Engine?
How is KV-cache implemented in autoregressive transformer inference?
What is continuous batching and how is it better than static batching?
What is Lucene and what are its key structures?
What is continuous batching and how is it implemented in vLLM/TGI?