Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

SPLICE: Latent Diffusion over JEPA Embeddings for Conformal Time-Series Inpainting

About

Generative models for time-series imputation achieve strong reconstruction accuracy, yet provide no finite-sample reliability guarantees, a critical limitation in power systems where imputed values inform dispatch and planning. We introduce SPLICE (Self-supervised Predictive Latent Inpainting with Conformal Envelopes), a modular framework coupling latent generative imputation with distribution-free, online-adaptive prediction intervals. A JEPA encoder maps daily load segments into a 64-dimensional latent space; a conditional latent bridge with four sampling modes generates candidate gap trajectories; an hourly-conditioned decoder maps back to signal space; and Adaptive Conformal Inference (ACI) wraps the output with coverage-guaranteed prediction bands. The flow-matching variant achieves comparable quality to DDIM in 5--10 ODE steps (5-10x speedup). On thirteen load datasets (nine proprietary, three UCI Electricity, ETTh1), SPLICE achieves the lowest mean Load-only MSE (0.056), winning 9/12 non-degenerate datasets at 91-day gaps and 18/32 across all gap lengths vs. five established baselines, and produces the best CRPS (0.161, -18.3% vs. the strongest competitor). ACI delivers 93--95% empirical coverage, correcting under-coverage failures of up to 7.5 pp observed with static conformal prediction. A pooled JEPA encoder trained on nine feeds transfers to four unseen domains, matching or exceeding per-dataset oracles with only a quick bridge fine-tuning.

Arnaud Zinflou• 2026

Related benchmarks

TaskDatasetResultRank
Time Series ImputationETTh1
MSE0.022
162
Time Series ImputationPepcoCOM 30-day gap
MSE0.0031
8
Time Series ImputationSEMAResNstar1009 30-day gap
MSE0.00e+0
8
Time Series ImputationUES_NH_Med 30-day gap
MSE0.5312
8
Time Series ImputationWCMA1010res 30-day gap
MSE0.0073
8
Time Series ImputationWCMAnatGridRes1004 30-day gap
MSE0.0064
8
Time Series ImputationUCI Elec MT 320 30-day gap
MSE0.017
8
Time Series ImputationETTh1 30-day gap
MSE0.0276
8
Time Series ImputationPepcoCOM min-max [0, 1] (7-day gap)
MSE0.0058
8
Time Series ImputationRICom1013 7-day gap min-max [0, 1]
MSE0.0864
8
Showing 10 of 73 rows
...

Other info

Follow for update