Training-Data on The Kiseki Log https://ki-seki.github.io/tags/training-data/ Recent content in Training-Data on The Kiseki Log Hugo -- 0.146.0 en 2023-2026 Elliot Shichao Song CC BY-SA 4.0 Sun, 25 Oct 2026 22:30:44 +0800 A Billion-Scale Retrieval System for Text-to-Image Training Data https://ki-seki.github.io/posts/261025-training-data-retrieval/ Sun, 25 Oct 2026 20:12:06 +0800 https://ki-seki.github.io/posts/261025-training-data-retrieval/ Training a text-to-image model from scratch means finding, inspecting and filtering samples across billions of image–text pairs. Using a retrieval system over 1.4B samples as an example, this post covers data storage and management, distributed embedding extraction, building and querying an IVF-PQ index, and online retrieval and cluster browsing.