ASH achieves state-of-the-art ANN recall and speed across compression levels by learning an orthonormal projection for dimensionality reduction followed by scalar quantization in an asymmetric encoder-decoder setup.
Locally-adaptive quantization for streaming vector search, February 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ASH: Asymmetric Scalar Hashing With Learned Dimensionality Reduction for High-Fidelity Vector Quantization
ASH achieves state-of-the-art ANN recall and speed across compression levels by learning an orthonormal projection for dimensionality reduction followed by scalar quantization in an asymmetric encoder-decoder setup.