SDFlow: Similarity-Driven Flow Matching for Time Series Generation

TL;DR

SDFlow introduces a non-autoregressive, flow matching framework operating in VQ latent space for efficient and high-quality long-sequence time series generation, overcoming exposure bias and improving inference speed.

Contribution

It presents a novel non-autoregressive approach with global transport mapping and manifold decomposition, achieving state-of-the-art results in time series generation.

Findings

SDFlow outperforms autoregressive models on key metrics.

It significantly reduces inference time for long sequences.

The method maintains high fidelity in generated time series.

Abstract

Vector quantization (VQ) with autoregressive (AR) token modeling is a widely adopted and highly competitive paradigm for time-series generation. However, such models are fundamentally limited by exposure bias: during inference, errors can accumulate across sequential predictions, leading to pronounced quality degradation in long-horizon generation. To address this, we propose SDFlow ($\textbf{S}$imilarity-$\textbf{D}$riven $\textbf{Flow}$ Matching), a non-autoregressive framework that operates entirely in the frozen VQ latent space and enables parallel sequence generation via flow matching. We tackle three key challenges in making this transition: (1) eliminating exposure bias by replacing step-wise token prediction with a global transport map; (2) mitigating the high-dimensionality of VQ token spaces via a low-rank manifold decomposition with a learned anchor prior over the latent manifold; and (3) incorporating discrete supervision into continuous transport dynamics by introducing a categorical posterior over codebook indices within a variational flow-matching formulation. Extensive experiments show that SDFlow achieves state-of-the-art performance, improving Discriminative Score and substantially reducing Context-FID, particularly for challenging long-sequence generation. Moreover, SDFlow provides significant inference speedups over autoregressive baselines, offering both high fidelity and computational efficiency. Code is available at https://anonymous.4open.science/r/SDFlow-D6F3/