Predicting Weak Retrieval Without an LLM
Cheap signals flag weak retrieval before you pay for a rerank or LLM judge. The one that works depends on how your search fails.
Dylan Couzon
June 24, 2026
Relevance & Evaluation
Understand relevance feedback, discovery, evaluation, and dataset quality.
Relevance Feedback in Qdrant
The story behind the vector search-native relevance feedback feature, available since 1.17.0, which increases the relevance of search results universally, cheaply, and at scale.
Evgeniya Sukhodolskaya
February 20, 2026
Relevance Feedback in Information Retrieval
A survey of relevance feedback methods from lexical and neural retrieval research, and why vector search engines lack them.
Evgeniya Sukhodolskaya
March 27, 2025
Discovery Search in Qdrant
Discover points by constraining the vector space.
Luis Cossío
January 31, 2024
Detecting Dataset Errors with Similarity Search
Find mislabeled items in a dataset by comparing category names with text and image embeddings.
George Panchuk
July 18, 2022