Generative Unordered Flow for Set-Structured Data Generation

TL;DR

This paper introduces unordered flow, a novel flow-based generative model tailored for set-structured data, converting unordered data into function representations and employing flow matching and particle filtering techniques, demonstrating superior performance on real-world datasets.

Contribution

The paper proposes unordered flow, a new method for generating set-structured data by transforming unordered data into functions and applying flow matching with particle filtering, addressing a gap in existing models.

Findings

Outperforms previous baselines in generating set-structured data.

Effectively models unordered data through function representations.

Demonstrates strong results on multiple real-world datasets.

Abstract

Flow-based generative models have demonstrated promising performance across a broad spectrum of data modalities (e.g., image and text). However, there are few works exploring their extension to unordered data (e.g., spatial point set), which is not trivial because previous models are mostly designed for vector data that are naturally ordered. In this paper, we present unordered flow, a type of flow-based generative model for set-structured data generation. Specifically, we convert unordered data into an appropriate function representation, and learn the probability measure of such representations through function-valued flow matching. For the inverse map from a function representation to unordered data, we propose a method similar to particle filtering, with Langevin dynamics to first warm-up the initial particles and gradient-based search to update them until convergence. We have conducted extensive experiments on multiple real-world datasets, showing that our unordered flow model is very effective in generating set-structured data and significantly outperforms previous baselines.