paper-with-me

홈 › Papers

Calibrated Test-Time Guidance for Bayesian Inference

2026-02-25 · Daniel Geyfman, Felix Draxler, Jan Groeneveld, Hyunsoo Lee, Theofanis Karaletsos, Stephan Mandt arxiv

Test-time guidance is a widely used mechanism for steering pretrained diffusion models toward outcomes specified by a reward function. Existing approaches, however, focus on maximizing reward rather than sampling from the true Bayesian posterior, leading to miscalibrated inference. In this work, we show that common test-time guidance methods do not recover the correct posterior distribution and identify the structural approximations responsible for this failure. We then propose consistent alternative estimators that enable calibrated sampling from the Bayesian posterior. We significantly outperform previous methods on a set of Bayesian inference tasks, and set a new state-of-the-art PSNR in black hole image reconstruction.

📄 PDF Abstract BibTeX arXiv:2602.22428

Code (0)

등록된 구현이 없습니다.

Tasks

Image ReconstructionBayesian Inference

Similar Papers 제목 키워드 기반

Testing for Causality in Continuous Time Bayesian Network Models of High-Frequency Data

2016-01-25 · Jonas Hallgren, Timo Koski

Continuous time Bayesian networks are investigated with a special focus on their ability to express causality. A framework is presented for doing inference in these networks. The central contributions are a representatio…

LRT-Diffusion: Calibrated Risk-Aware Guidance for Diffusion Policies

2025-10-28 · Ximan Sun, Xiang Cheng arxiv

Diffusion policies are competitive for offline reinforcement learning (RL) but are typically guided at sampling time by heuristics that lack a statistical notion of risk. We introduce LRT-Diffusion, a risk-aware sampling…

Reinforcement LearningOffline RL

Low-Budget Simulation-Based Inference with Bayesian Neural Networks

2024-08-27 · Arnaud Delaunoy, Maxence de la Brassinne Bonardeaux, Siddharth Mishra-Sharma, Gilles Louppe

Simulation-based inference methods have been shown to be inaccurate in the data-poor regime, when training simulations are limited or expensive. Under these circumstances, the inference network is particularly prone to o…

Probabilistic Meta-Learning for Bayesian Optimization

2021-01-01 · Felix Berkenkamp, Anna Eivazi, Lukas Grossberger, Kathrin Skubch 외

Transfer and meta-learning algorithms leverage evaluations on related tasks in order to significantly speed up learning or optimization on a new problem. For applications that depend on uncertainty estimates, e.g., in Ba…

Bayesian OptimizationMeta-LearningTransfer Learning

Bayesian Inference for Large Scale Image Classification

2019-08-09 · Jonathan Heek, Nal Kalchbrenner

Bayesian inference promises to ground and improve the performance of deep neural networks. It promises to be robust to overfitting, to simplify the training procedure and the space of hyperparameters, and to provide a ca…

Bayesian InferenceClassificationDecision MakingFairness+3