Free-form flows: Make any architecture a normalizing flow

F Draxler, P Sorrenson… - International …, 2024 - proceedings.mlr.press
Normalizing Flows are generative models that directly maximize the likelihood. Previously,
the design of normalizing flows was largely constrained by the need for analytical …

Steering masked discrete diffusion models via discrete denoising posterior prediction

J Rector-Brooks, M Hasan, Z Peng, Z Quinn… - arxiv preprint arxiv …, 2024 - arxiv.org
Generative modeling of discrete data underlies important applications spanning text-based
agents like ChatGPT to the design of the very building blocks of life in protein sequences …

All-in-one simulation-based inference

M Gloeckler, M Deistler, C Weilbach, F Wood… - arxiv preprint arxiv …, 2024 - arxiv.org
Amortized Bayesian inference trains neural networks to solve stochastic inference problems
using model simulations, thereby making it possible to rapidly perform Bayesian inference …

Compositional simulation-based inference for time series

M Gloeckler, S Toyota, K Fukumizu… - arxiv preprint arxiv …, 2024 - arxiv.org
Amortized simulation-based inference (SBI) methods train neural networks on simulated
data to perform Bayesian inference. While this approach avoids the need for tractable …

Variational Flow Matching for Graph Generation

F Eijkelboom, G Bartosh, CA Naesseth… - arxiv preprint arxiv …, 2024 - arxiv.org
We present a formulation of flow matching as variational inference, which we refer to as
variational flow matching (VFM). Based on this formulation we develop CatFlow, a flow …

A kernel-based conditional two-sample test using nearest neighbors (with applications to calibration, regression curves, and simulation-based inference)

A Chatterjee, Z Niu, BB Bhattacharya - arxiv preprint arxiv:2407.16550, 2024 - arxiv.org
In this paper we introduce a kernel-based measure for detecting differences between two
conditional distributions. Using thekernel trick'and nearest-neighbor graphs, we propose a …

Synthesizing 3D Abstractions by Inverting Procedural Buildings with Transformers

M Dax, J Berbel, J Stria, L Guibas… - arxiv preprint arxiv …, 2025 - arxiv.org
We generate abstractions of buildings, reflecting the essential aspects of their geometry and
structure, by learning to invert procedural models. We first build a dataset of abstract …

SGFM: Conditional Flow Matching for Time Series Anomaly Detection With State Space Models

Y He, T Yan, Y Zhan, Z Feng… - IEEE Internet of Things …, 2024 - ieeexplore.ieee.org
The industrial Internet of Things (IoT) landscape is enriched with a diverse array of sensors,
which are configured for the real-time monitoring and data collection to improve the …

Amortized Bayesian Multilevel Models

D Habermann, M Schmitt, L Kühmichel… - arxiv preprint arxiv …, 2024 - arxiv.org
Multilevel models (MLMs) are a central building block of the Bayesian workflow. They
enable joint, interpretable modeling of data across hierarchical levels and provide a fully …

The Unreasonable Effectiveness of Solving Inverse Problems with Neural Networks

P Holl, N Thuerey - arxiv preprint arxiv:2408.08119, 2024 - arxiv.org
Finding model parameters from data is an essential task in science and engineering, from
weather and climate forecasts to plasma control. Previous works have employed neural …