# Flash-SD-KDE: Accelerating SD-KDE with Tensor Cores

arXiv: https://arxiv.org/abs/2602.10378 · HTML: https://mlsanity.com/paper/2602.10378/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 26 references that are in the map.
- First posted on arXiv 10 Feb 2026.

## Builds on (most cited first, 8 of 8)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Generative Modeling by Estimating Gradients of the Data Distribution](https://mlsanity.com/paper/1907.05600/) (2019)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Mixed Precision Training](https://mlsanity.com/paper/1710.03740/) (2017)
- [Kernel Operations on the GPU, with Autodiff, without Memory Overflows](https://mlsanity.com/paper/2004.11127/) (2020)
- [Gradient Estimators for Implicit Models](https://mlsanity.com/paper/1705.07107/) (2017)
- [SD-KDE: Score-Debiased Kernel Density Estimation](https://mlsanity.com/paper/2504.19084/) (2025)

## Built on by (most cited first, 1 of 1)

- [Domain-Filtered Knowledge Graphs from Sparse Autoencoder Features](https://mlsanity.com/paper/2604.23829/) (2026)

## Abstract (arXiv)

Score-debiased kernel density estimation (SD-KDE) achieves improved asymptotic convergence rates over classical KDE, but its use of an empirical score has made it significantly slower in practice. We show that by re-ordering the SD-KDE computation to expose matrix-multiplication structure, Tensor Cores can be used to accelerate the GPU implementation. On a 32k-sample 16-dimensional problem, our approach runs up to 47× faster than a strong SD-KDE GPU baseline and 3,300× faster than scikit-learn's KDE. On a larger 1M-sample 16-dimensional task evaluated on 131k queries, Flash-SD-KDE completes in 2.3 s on a single GPU, making score-debiased density estimation practical at previously infeasible scales.