variational-inference

Tag

Cards List
#variational-inference

Modeling Document and Section Hierarchies in Embedding Space

arXiv cs.CL ↗ · 2d ago Cached

Columbia University researchers introduce the Hierarchical Embedded Topic Model (hETM), which couples document- and section-level topic distributions via a hierarchical prior in embedding space, with amortized variational inference showing improved perplexity, topic quality, and clustering across legal, financial, and forum corpora.

0 favorites 0 likes
#variational-inference

Amortized Localized Posterior Inference for Model Calibration with Application to an Ocean Turbulence Closure

arXiv cs.AI ↗ · 2026-10-31 Cached

提出了一种基于条件归一化流的摊销变分推断方法,构建加权似然的局部后验来处理模型校准中的模型偏差与上下文不确定性,并在Branin-Hoo孪生实验及单柱海洋模型的k-ε湍流闭合校准中验证了其相较于全局后验的误差降低效果。

0 favorites 0 likes
#variational-inference

VampPrior Unleashed

arXiv cs.LG ↗ · 2026-10-21 Cached

The paper identifies two key limitations of the VampPrior — tied prior/posterior covariances and fixed uniform mixture weights — and proposes decoupling covariances plus a Dirichlet process mixture, improving model evidence and clustering, notably when added to scVI across six single-cell atlases.

0 favorites 0 likes
#variational-inference

Large Language Bayes Is Not Reparameterisation-Invariant

arXiv cs.LG ↗ · 2026-10-03 Cached

This paper shows that Large Language Bayes (LLB), which averages posteriors from language-model-generated probabilistic programs using weights from an exponentiated evidence bound, is not reparameterisation-invariant. Equivalent program writings can receive substantially different weights (up to 31.9× discrepancy that can even reverse sign), inverting Bayes factors and inducing error in model posteriors controlled by the spread of bound shortfalls via a sharp Hilbert-distance bound.

0 favorites 0 likes
#variational-inference

Hierarchical Continuous Diffusion Language Models

Hugging Face Daily Papers ↗ · 2026-10-01 Cached

The paper proposes Hierarchical Continuous Diffusion Language Models (HC-DLM), which couple discrete token generation with a continuous latent trajectory in a single denoising process derived from a variational bound on token likelihood. HC-DLM improves over discrete and continuous diffusion baselines on Sudoku, Countdown, and LM1B tasks.

0 favorites 0 likes
#variational-inference

FB-GDM: Fully-Bayesian Guided Diffusion Models for High-Dimensional Linear Inverse Problems via Unsupervised Variational Inference

arXiv cs.LG ↗ · 2026-09-25 Cached

FB-GDM introduces a fully-Bayesian guided diffusion method for high-dimensional linear inverse problems, eliminating per-task hyperparameter tuning and demonstrating robust performance improvements over existing techniques.

0 favorites 0 likes
#variational-inference

RiVaT-Fuse: Reliability-Calibrated Variational Tensor Fusion for Multimodal Prediction under Modality Uncertainty

arXiv cs.LG ↗ · 2026-09-11 Cached

This paper proposes RiVaT-Fuse, a reliability-calibrated variational tensor fusion framework for multimodal prediction under modality uncertainty, which estimates a sample-wise latent state to improve predictive performance and stability in image–metadata tasks.

0 favorites 0 likes
#variational-inference

SQS: Bayesian DNN Compression through Sparse Quantized Sub-distributions

Hugging Face Daily Papers ↗ · 2026-09-07 Cached

The paper introduces a unified Bayesian variational framework combining spike-and-slab sparsity and Gaussian mixture quantization for high compression rates in large neural networks with minimal accuracy loss.

0 favorites 0 likes
#variational-inference

Expected Free Energy as Belief-Dependent Utility for rho-POMDPs

arXiv cs.AI ↗ · 2026-07-21 Cached

Shows that minimizing Expected Free Energy (EFE) is equivalent to solving a ρ-POMDP with utility as expected information gain, with exploration weight fixed at 1. Proves equivalence for observe-then-commit POMDPs and extends to factored observation POMDPs, with experiments demonstrating untuned weight matches or outperforms reward-only planning.

0 favorites 0 likes
#variational-inference

Uncovering Latent Reasoning Strategies in Language Models

Hugging Face Daily Papers ↗ · 2026-07-20 Cached

This paper proposes a method to decompose the response distribution of language models into structured, strategy-conditioned representations using a latent variable, addressing posterior collapse with a model-directed reconstruction objective.

0 favorites 0 likes
#variational-inference

Recovering Latent Structures after Variational Bayesian Variable Selection: Fit Assessment and Factor-Number Selection in Partially Exploratory Factor Analysis

arXiv cs.CL ↗ · 2026-07-09 Cached

This paper develops post-selection fit assessment for partially exploratory factor analysis (PEFA) under variational Bayesian variable selection, proposing absolute and relative fit diagnostics and a scale-free gain rule for factor-number selection, validated by simulations and an empirical example.

0 favorites 0 likes
#variational-inference

@KirkDBorne: "Pen and Paper Exercises in Machine Learning" Download 211-page PDF: http://arxiv.org/abs/2206.13446 Author’s GitHub: h…

X AI KOLs Timeline ↗ · 2026-07-03 Cached

This paper provides a 211-page collection of pen-and-paper exercises covering key topics in machine learning, including linear algebra, optimisation, graphical models, and variational inference, intended as an educational resource.

0 favorites 0 likes
#variational-inference

Equivariance and Augmentation for Bayesian Neural Networks

arXiv cs.LG ↗ · 2026-06-26 Cached

This paper studies data augmentation for Bayesian neural networks trained with variational inference, deriving conditions for exact equivariance and introducing novel symmetrization techniques like orbit expansion to improve symmetry and performance.

0 favorites 0 likes
#variational-inference

Fast and Slow Variational Continual Learning

arXiv cs.LG ↗ · 2026-06-24 Cached

This paper introduces the Continual IVON (CoVON) optimizer, which integrates fast and slow adaptation into variational continual learning to balance stability and plasticity, outperforming existing methods in domain-incremental learning, continual pre-training, and fine-tuning of large language models.

0 favorites 0 likes
#variational-inference

Implicit Variational Rejection Sampling

arXiv cs.LG ↗ · 2026-06-15 Cached

The article proposes Implicit Variational Rejection Sampling (IVRS), which integrates implicit distributions with rejection sampling to improve posterior approximation in variational inference, and introduces the Implicit Resampling Evidence Lower Bound (IR-ELBO) as a tighter variational lower bound.

0 favorites 0 likes
#variational-inference

What Type of Inference is Active Inference?

arXiv cs.AI ↗ · 2026-06-04 Cached

This paper analyzes Active Inference by proving that the Variational Free Energy of an augmented generative model can be decomposed into the predictive model's VFE plus explicit entropy-correction terms, yielding a full variational characterization of Expected Free Energy-based planning. The authors derive a message-passing scheme for EFE-based planning and validate it on grid-world environments.

0 favorites 0 likes
#variational-inference

Variational Inference for Evidential Deep Learning

arXiv cs.LG ↗ · 2026-05-27 Cached

A mathematically principled framework, Variational Inference Evidential Deep Learning (VI-EDL), is proposed to address limitations in conventional Evidential Deep Learning by reformulating it through variational inference, deriving an Evidence Lower Bound, establishing a generalization bound, and achieving state-of-the-art performance on visual and medical datasets.

0 favorites 0 likes
#variational-inference

Amortized Factor Inference Networks for Posterior Inference

arXiv cs.LG ↗ · 2026-05-27 Cached

Introduces Amortized Factor Inference Networks (AFINs), a family of encode-merge-decode inference networks that generalize across varying priors, likelihoods, and dimensionality, achieving posterior accuracy comparable to NUTS with much less compute.

0 favorites 0 likes
#variational-inference

Three Costs of Amortizing Gaussian Process Inference with Neural Processes

arXiv cs.LG ↗ · 2026-05-22 Cached

This paper decomposes the predictive KL divergence between Gaussian process and latent neural process posteriors into three terms, providing upper bounds that characterize approximation errors and connecting representation dimension to kernel smoothness.

0 favorites 0 likes
#variational-inference

Closed-form predictive coding via hierarchical Gaussian filters

arXiv cs.LG ↗ · 2026-05-21 Cached

The paper introduces closed-form predictive coding via hierarchical Gaussian filters that restore precision-weighted prediction errors, yielding faster and more efficient training without global error signals, outperforming backpropagation on certain tasks.

0 favorites 0 likes
Next →
← Back to home

Submit Feedback