How to Unlock Time Series Editing? Diffusion-Driven Approach with Multi-Grained Control

TL;DR

This paper introduces CocktailEdit, a diffusion-based framework that enables precise, flexible editing of time series data by controlling point-level and segment-level properties simultaneously, maintaining temporal coherence.

Contribution

It presents a novel multi-grained control framework for time series editing that combines anchor and classifier-based controls within diffusion models.

Findings

Effective local control during denoising inference.

Maintains temporal coherence in edited sequences.

Works seamlessly with various diffusion-based models.

Abstract

Recent advances in time series generation have shown promise, yet controlling properties in generated sequences remains challenging. Time Series Editing (TSE) - making precise modifications while preserving temporal coherence - consider both point-level constraints and segment-level controls that current methods struggle to provide. We introduce the CocktailEdit framework to enable simultaneous, flexible control across different types of constraints. This framework combines two key mechanisms: a confidence-weighted anchor control for point-wise constraints and a classifier-based control for managing statistical properties such as sums and averages over segments. Our methods achieve precise local control during the denoising inference stage while maintaining temporal coherence and integrating seamlessly, with any conditionally trained diffusion-based time series models. Extensive experiments across diverse datasets and models demonstrate its effectiveness. Our work bridges the gap between pure generative modeling and real-world time series editing needs, offering a flexible solution for human-in-the-loop time series generation and editing. The code and demo are provided for validation.