Research map / Optimization / Optimizers
Multiscale Quantization for Fast Similarity Search
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add scalar quantization of residual norm scales and end-to-end training on top of learned rotation (ablation) | IVFADC with rotation · SIFT1M | MSE reduction: additional 10-15% | Helped |
| add scalar quantization of residual norm scales and end-to-end training on top of learned rotation (ablation) | IVFADC with rotation · DEEP10M | MSE reduction: additional 5-10% | Helped |
| MSQ versus PQ, OPQ and stacked quantizers at fixed bitrate | MSQ · SIFT1M and DEEP10M | Recall1@N | Helped |
| use 8 scalar quantizers for norm scales instead of unquantized scales (ablation) | MSQ · SIFT1M and DEEP10M | recall | No clear effect |
| use 16-center in-register LUT instead of 256-center in-memory LUT | ADC search · large neighbor retrieval | Recall1@N and speed: 5x speed up | Helped |
Closest papers by meaning
- Norm-Explicit Quantization: Improving Vector Quantization for Maximum Inner Product Search (AAAI 2020 · similarity 0.90)
- Supervised Quantization for Similarity Search (CVPR 2016 · similarity 0.89)
- Accelerating Large-Scale Inference with Anisotropic Vector Quantization (ICML 2020 · similarity 0.88)
- HiHPQ: Hierarchical Hyperbolic Product Quantization for Unsupervised Image Retrieval (AAAI 2024 · similarity 0.86)
- Qinco2: Vector Compression and Search with Improved Implicit Neural Codebooks (ICLR 2025 · similarity 0.85)
- Optimal and Approximate Adaptive Stochastic Quantization (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.