arXiv Machine Learning
Beyond Local Accuracy: Benchmarking the Rollout Reliability of PDE Foundation Models
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Machine Learning
Aug 4
Round-Trip Consistency: Bidirectional Diffusion Models Can Predict Their Own Rollout Errors
arXiv:2608. 00675v1 Announce Type: cross Abstract: Autoregressive models accumulate error over long rollouts, yet at deployment there is no ground truth to measure it against.
arXiv Machine Learning
Jul 7
When Do Foundation Models Pay Off? A Break-Even Analysis of Pretrained Time Series Forecasters
arXiv:2607. 04919v1 Announce Type: new Abstract: Deploying a time series foundation model requires GPU infrastructure, engineering overhead, and carries no guarantee of improvement over XGBoost.
arXiv AI
Aug 3
HERO: History-Enriched Rollout Training for Long-Horizon Autoregressive Neural Operators
arXiv:2607. 29135v1 Announce Type: cross Abstract: Neural operators provide fast surrogates for time-dependent partial differential equations (PDEs) by applying a learned evolution operator recursively to its own predictions, but this autoregressive rollout feeds every prediction error back as input, so local errors accumulate.
arXiv Machine Learning
Sep 30
PDE-OBS: Controlled Evaluation Across Observation Patterns
arXiv:2609.36521v1 Announce Type: new Abstract: Physical-field reconstruction and forecasting depend on both measurement density and spatial layout, yet evaluation under a single observation pattern...
arXiv AI
Jul 8
The Granularity Paradox: How Temporal Disaggregation Inflates In-Sample Fit and Compounds Out-of-Sample Error
arXiv:2607. 05450v1 Announce Type: cross Abstract: This paper explores the "Granularity Paradox" in time-series forecasting, wherein finer temporal disaggregation (e.
arXiv Machine Learning
Sep 10
Cadence: Error-Bounded Lossy Compression of Demand Time Series with a Time-Series Foundation Model
arXiv:2609.06008v1 Announce Type: cross Abstract: We present Cadence, an error-bounded lossy compressor for numeric time series pairing a 330M-parameter time-series foundation model (Google TimesFM-3...