How does stemming differ from lemmatization? In which tasks should each be used?
Machine Learning / AI
Why is BF16 more convenient than FP16 when training large models?
What checks should be made when choosing a learning rate for SFT LLM?
Why is gradient boosting more commonly used in industrial environments than random forests?
What are Delta Lake / Iceberg / Hudi and why lakehouse?
What is multimodal RAG and how to create an index for images?
What popular AutoML solutions are there (H2O, AutoGluon, FLAML, AutoKeras)?
Have you worked with Apache Spark?
What are the modern trends in training on mixed GPU+CPU+TPU clusters?
What is a bounding box and what are the coordinate formats (xyxy, xywh, cxcywh)?
What is LambdaRank?
What methods are used for anomaly detection in images (PaDiM, PatchCore)?
What is MCMC and what methods are there (Metropolis-Hastings, Gibbs, HMC)?
What are anchor-free detectors? Tell us about FCOS.
What is OWL-ViT?
What is ROUGE and how does ROUGE-L differ from ROUGE-N?
What is boundary loss and why is it needed?
What tasks do you consider ideal and what do you want to do?
What are knowledge graph embeddings (TransE, RotatE, ComplEx)?
What is async checkpointing and how does it work?