AI Paper Atlas

Research map / Time series / Causal and generative time series

Latent-to-Data Cascaded Diffusion Models for Unconditional Time Series Generation

Lifeng Shen, Kai Syun Hou, Weiyu Chen, James Kwok

ICLR 2026 · Time series › Causal and generative time series · approach: Diffusion models for time series · 0 citations (OpenAlex)

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
add data-space diffusion after latent diffusion (ablation)L2D-Diff latent-only · StockC-FID: 3.682 -> 0.310Improved
add data-space diffusion after latent diffusion (ablation)L2D-Diff latent-only · Character TrajectoriesC-FID: 1.829 -> 0.284Improved
add latent-space conditioning to data-space diffusion (ablation)L2D-Diff data-only · StockC-FID: 0.385 -> 0.310Improved
add latent-space conditioning to data-space diffusion (ablation)L2D-Diff data-only · Character TrajectoriesC-FID: 2.368 -> 0.284Improved
use CNN instead of MLP for conditioning network F (ablation)L2D-Diff · Character TrajectoriesC-FID: 0.43 -> 0.28Improved
use depth 5 instead of 1 layer in conditioning network (ablation)L2D-Diff · Character TrajectoriesC-FID: 0.58 -> 0.28Improved
increase conditioning network depth from 5 to 20 layers (ablation)L2D-Diff · StocksC-FID: 0.31 -> 0.75Hurt
use concatenation instead of cross-attention for latent conditioning (ablation)L2D-Diff · Japanese Vowels / Character TrajectoriesC-FID: 0.61 -> 0.28Improved
use concatenation instead of direct addition for latent conditioning (ablation)L2D-Diff · StocksC-FID: 0.55 -> 0.31Improved
use concatenation instead of FiLM (ablation)L2D-Diff · StocksC-FID: 0.32 -> 0.31No clear effect
increase latent dimension from 8 to 64 (ablation)L2D-Diff · StockC-FID: 0.310 -> 1.121Hurt
increase latent dimension from 8 to 128 (ablation)L2D-Diff · Character TrajectoriesC-FID: 0.284 -> 2.947Hurt

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