miniCOIL: on the Road to Usable Sparse Neural Retrieval
Introducing miniCOIL, a lightweight sparse neural retriever capable of generalization.
Evgeniya Sukhodolskaya
May 13, 2025
Embedding Research
Explore the research behind modern embeddings and neural retrieval. Dive into sparse neural models, late interaction, metric learning, and new baselines for hybrid search.
Modern Sparse Neural Retrieval: From Theory to Practice
A comprehensive guide to modern sparse neural retrievers: COIL, TILDEv2, SPLADE, and more. Find out how they work and learn how to use them effectively.
Evgeniya Sukhodolskaya
October 23, 2024
Late Interaction Retrieval with Dense Token Embeddings
We recently discovered that embedding models can become late interaction models & can perform surprisingly well in some scenarios. See what we learned here.
Kacper Łukawski
August 14, 2024
BM42: Attention-Based Sparse Embeddings for Hybrid Search
Introducing BM42 - a new sparse embedding approach, which combines the benefits of exact keyword search with the intelligence of transformers.
Andrey Vasnetsov
July 01, 2024
Layer Recycling and Fine-tuning Efficiency
Learn when and how to use layer recycling to achieve different performance targets.
Yusuf Sarıgöz
August 23, 2022
Fine Tuning Similar Cars Search
Learn how to train a similarity model that can retrieve similar car images in novel categories.
Yusuf Sarıgöz
June 28, 2022
Metric Learning for Anomaly Detection
Practical use of metric learning for anomaly detection. A way to match the results of a classification-based approach with only ~0.6% of the labeled data.
Yusuf Sarıgöz
May 04, 2022
Advanced Introduction to Triplet Loss
What are the advantages of Triplet Loss over Contrastive loss and how to efficiently implement it?
Yusuf Sarıgöz
March 24, 2022
Metric Learning Tips & Tricks
Practical recommendations on how to train a matching model and serve it in production. Even with no labeled data.
Andrei Vasnetsov
May 15, 2021