Cross-View Time-Frequency Diffusion for Extreme-Aware Time-Series Generation
We propose Cross-View Time-Frequency Diffusion (CrossTF), a framework that exploits discrepancies between time- and frequency-view denoisers for extreme-aware time-series generation.
ProofPaper ↗
Key points
- Generating realistic time-series requires faithfully reproducing rare but high-impact extreme events.
- Our key empirical finding is extreme-sensitive cross-view asymmetry: moving along the direction connecting the two views' clean-sample estimates substantially changes extreme fidelity while having limited effects on overall generation quality.
- These findings motivate a staged optimization strategy in CrossTF: after training and freezing both denoisers, it only needs to learn step-specific fusion parameters within this window using an Extreme Value Theory-based objective under an energy-distance constraint.
- These results highlight cross-view discrepancy as an effective basis for low-dimensional optimization in extreme-aware time-series generation.
Sources (1)
- [1]Cross-View Time-Frequency Diffusion for Extreme-Aware Time-Series GenerationarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 12, 09:13 PM
We propose Cross-View Time-Frequency Diffusion (CrossTF), a framework that exploits discrepancies between time- and frequency-view denoisers for extreme-aware time-series generation.
Generating realistic time-series requires faithfully reproducing rare but high-impact extreme events.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 12, 2026From Conceptual EneMonIE to an Executable Backend: A Modular Architecture for Energy Information Extractionand Grounded Question Answering
- Oct 11, 2026AesCode: Aesthetic Code Generation with Decoupled Cross-Modal Rewards
- Oct 11, 2026WUSPruner: Second-Moment-Optimal Transforms for Structured Pruning
- Oct 11, 2026Apertus 1.5 Release Notes
- Oct 10, 2026vllm-sr/d3
- Oct 10, 2026CAG-WAM: Causal Action-Gain World Action Models for Robot Manipulation