{"repo":"VectorDB-NTU/Extended-RaBitQ","free":true,"listed":false,"github":"https://github.com/VectorDB-NTU/Extended-RaBitQ","clone":"git clone https://github.com/VectorDB-NTU/Extended-RaBitQ.git","description":"The repo has been moved to https://github.com/VectorDB-NTU/RaBitQ-Library. [SIGMOD 2025] Practical and Asymptotically Optimal Quantization of High-Dimensional Vectors in Euclidean Space for Approximate Nearest Neighbor Search","language":"C++","stars":72,"topics":["nearest-neighbor-search","quantization","vector-database"],"license":"Apache-2.0","category":"databases-storage","readme_excerpt":"Extended RaBitQ The repo has been archived. Please refer to the RaBitQ-Library for further development. News: A library with more practical implementation techniques about RaBitQ is released at the RaBitQ-Library. News: The paper (arXiv:2409.09913, September, 2024) has been accepted by SIGMOD 2025. [SIGMOD 2025] Practical and Asymptotically Optimal Quantization of High-Dimensional Vectors in Euclidean Space for Approximate Nearest Neighbor Search --- Replace your scalar and binary quantization with RaBitQ seamlessly. Enjoy blazingly fast distance computation with dominant accuracy. The project proposes a novel quantization algorithm developped from RaBitQ. The algorithm supports to compress high-dimensional vectors with arbitrary compression rates. Its computation is exactly the same as the classical scalar quantization and has dominant accuracy under same compression rates. It brings especially significant improvement in the setting from 2-bit to 6-bit, which helps an algorithm to achieve high recall without reranking. We summarize the key intuitions and results as follows. For more details, please refer to our paper https://arxiv.org/pdf/2409.09913. Prepapring Prerequisites Please refer to ./inc/third/README.md for detailed information about third-party libraries. AVX512 is required Compiling Source codes are stored in ./src , binary files are stored in ./bin please update the cmake file in ./src after adding new source files. Datasets Download and preprocess the datasets. ","default_branch":null,"files":null,"tree":[],"storefront":"/r/VectorDB-NTU","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/VectorDB-NTU/Extended-RaBitQ/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}