AI Paper Atlas

Research map / Time series / Causal and generative time series

Primal-Spectral Generative Modeling: Fast Analytical Generation via Pseudoinverse Lévy Inversion

Zhiyuan Ouyang, Simei Huang, Zuokai Wen, Xiangyun Zhang, Junchi Yan

ICML 2026 · Time series › Causal and generative time series · approach: Irregular series and imputation

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (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 changedWhereMeasuredEffect
use PriSpecNet-TS 50-NFE Heun instead of ARMDtime series generation · ETThContext-FID: 0.093 -> 0.018Improved
use PriSpecNet-TS 50-NFE Heun instead of ARMDtime series generation · StockContext-FID: 0.141 -> 0.052Improved
use PriSpecNet-TS 50-NFE Heun instead of ARMDforecasting · SolarMSE: 0.161 -> 0.113Improved
use PriSpecNet-TS 50-NFE Heun instead of ARMDforecasting · StockMSE: 0.235 -> 0.179Improved
use 1-NFE PiLI instead of 50-NFE Heun sampler (ablation)PriSpecNet-TS · ETThContext-FID: 0.018 -> 0.033Hurt
use 1-NFE PiLI instead of 25-NFE DPM-Solver++ (ablation)PriSpecNet-XL · ImageNet 256x256FID: 1.62 -> 1.66Hurt
use 1-NFE PiLI instead of 25-NFE DPM-Solver++ (ablation)PriSpecNet-XL · ImageNet 256x256Total Gflops: 4,436 -> 26Improved
use PiLI-XL instead of iMF-XL/21-NFE generation · ImageNet 256x256FID: 1.72 -> 1.66Improved
add spectral components with K=256 to PriSpecNet-B (early training) (ablation)PriSpecNet-B · ImageNet 256x256 (200k steps)FID: 128.53 -> 118.64Improved
increase spectral dimension K from 256 to 768 (ablation)PriSpecNet-B · ImageNet 256x256 (final)FID: 98.09 -> 48.00Improved
increase spectral dimension K from 768 to 1280 (ablation)PriSpecNet-B · ImageNet 256x256FID: 48.00 -> above 140Hurt
increase spectral dimension K from 768 to 1280 (ablation)PriSpecNet-XL · ImageNet 256x256 (final)FID: 14.53 -> 1.62Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy