JET is a conditional flow matching framework that generates EEG as continuous raw sequences with added constraints for spectral and temporal properties, achieving over 40% lower TS-FID than prior discrete denoising methods on three benchmarks.
Title resolution pending
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
representative citing papers
The first paired POCUS-to-high-end ultrasound dataset is released and a cGAN baseline raises SSIM from 0.29 to 0.54 and PSNR from 19.16 dB to 22.41 dB on 1064 test pairs.
K-DSM uses per-feature kurtosis to set noise scales in DSM, enabling effective single-scale anomaly detection on tabular benchmarks in both semi-supervised and unsupervised settings.
Gated SAEs decouple which features to use from how large their activations should be, applying the L1 penalty only to selection and thereby eliminating shrinkage while halving the number of firing features needed for good fidelity.
PiFM extends Flow Matching to multi-parameter settings by enforcing path-independent flows that approximate Wasserstein barycenters under suitable assumptions.
TabGRAA applies group-relative advantage alignment in an iterative reward-guided post-training loop to improve tabular language model generators on fidelity, utility, and privacy trade-offs across five benchmarks.
citing papers explorer
-
Let EEG Models Learn EEG
JET is a conditional flow matching framework that generates EEG as continuous raw sequences with added constraints for spectral and temporal properties, achieving over 40% lower TS-FID than prior discrete denoising methods on three benchmarks.
-
A Paired Point-of-Care Ultrasound Dataset for Image Quality Enhancement and Benchmarking via a cGAN Baseline
The first paired POCUS-to-high-end ultrasound dataset is released and a cGAN baseline raises SSIM from 0.29 to 0.54 and PSNR from 19.16 dB to 22.41 dB on 1064 test pairs.
-
Kurtosis-Guided Denoising Score Matching for Tabular Anomaly Detection
K-DSM uses per-feature kurtosis to set noise scales in DSM, enabling effective single-scale anomaly detection on tabular benchmarks in both semi-supervised and unsupervised settings.
-
Improving Dictionary Learning with Gated Sparse Autoencoders
Gated SAEs decouple which features to use from how large their activations should be, applying the L1 penalty only to selection and thereby eliminating shrinkage while halving the number of firing features needed for good fidelity.
-
Path-independent Flow Matching for Multi-parameter Generative Dynamics
PiFM extends Flow Matching to multi-parameter settings by enforcing path-independent flows that approximate Wasserstein barycenters under suitable assumptions.
-
Self-Improving Tabular Language Models via Iterative Reward-Guided Post-Training
TabGRAA applies group-relative advantage alignment in an iterative reward-guided post-training loop to improve tabular language model generators on fidelity, utility, and privacy trade-offs across five benchmarks.