Method
Embedding fine-tuning
Fine-tuning embedding models on domain triplets or synthetic queries.
26 cases from 21 companies apply Embedding fine-tuning. 7 were checked in an online A/B test; 4 report results in production.
Typical tasks: Rank search results, Retrieve and rank from millions of items, Learn user and item embeddings, Build an assistant over internal docs and 5 more. Companies: LinkedIn, Mercari, Wildberries, Atlassian, Criteo, OLX and 15 more. Often combined with: Dense retrieval, Contrastive learning, RAG, LoRA fine-tuning, Two-tower model and 59 more.





