Guy Van den Broeck

Affiliations

University of California, Los Angeles

papers

432

total citations

papers (20)

TRACE Back from the Future: A Probabilistic Reasoning Approach to Controllable Language Generation

ICML 2025

citations

Plug-and-Play Context Feature Reuse for Efficient Masked Generation

NEURIPS 2025arXiv

citations

Rao-Blackwell Gradient Estimators for Equivariant Denoising Diffusion

NEURIPS 2025arXiv

citations

Tractable Transformers for Flexible Conditional Generation

ICML 2025arXiv

citations

Probabilistic Inference with Algebraic Constraints: Theoretical Limits and Practical Approximations

NEURIPS 2020

citations

A Compositional Atlas of Tractable Circuit Operations for Probabilistic Inference

NEURIPS 2021

citations

A Unified Approach to Count-Based Weakly Supervised Learning

NEURIPS 2023

citations

Guy Van den Broeck

Affiliations

papers (20)

Semantic Probabilistic Layers for Neuro-Symbolic Learning

Counterexample-Guided Learning of Monotonic Neural Networks

Accelerating Diffusion LLMs via Adaptive Parallel Decoding

Tractable Regularization of Probabilistic Circuits

Image Inpainting via Tractable Steering of Diffusion Models

Probabilistically Rewired Message-Passing Neural Networks

Sparse Probabilistic Circuits via Pruning and Growing

Learning to Discretize Denoising Diffusion ODEs

A Pseudo-Semantic Loss for Autoregressive Models with Logical Constraints

Scaling Tractable Probabilistic Circuits: A Systems Perspective

On the Relationship Between Monotone and Squared Probabilistic Circuits

Collapsed Inference for Bayesian Deep Learning

Controllable Generation via Locally Constrained Resampling

TRACE Back from the Future: A Probabilistic Reasoning Approach to Controllable Language Generation

Plug-and-Play Context Feature Reuse for Efficient Masked Generation

Rao-Blackwell Gradient Estimators for Equivariant Denoising Diffusion

Tractable Transformers for Flexible Conditional Generation

Probabilistic Inference with Algebraic Constraints: Theoretical Limits and Practical Approximations

A Compositional Atlas of Tractable Circuit Operations for Probabilistic Inference

A Unified Approach to Count-Based Weakly Supervised Learning

papers (20)

Semantic Probabilistic Layers for Neuro-Symbolic Learning

Counterexample-Guided Learning of Monotonic Neural Networks

Accelerating Diffusion LLMs via Adaptive Parallel Decoding

Tractable Regularization of Probabilistic Circuits

Image Inpainting via Tractable Steering of Diffusion Models

Probabilistically Rewired Message-Passing Neural Networks

Sparse Probabilistic Circuits via Pruning and Growing

Learning to Discretize Denoising Diffusion ODEs

A Pseudo-Semantic Loss for Autoregressive Models with Logical Constraints

Scaling Tractable Probabilistic Circuits: A Systems Perspective

On the Relationship Between Monotone and Squared Probabilistic Circuits

Collapsed Inference for Bayesian Deep Learning

Controllable Generation via Locally Constrained Resampling

TRACE Back from the Future: A Probabilistic Reasoning Approach to Controllable Language Generation

Plug-and-Play Context Feature Reuse for Efficient Masked Generation

Rao-Blackwell Gradient Estimators for Equivariant Denoising Diffusion

Tractable Transformers for Flexible Conditional Generation

Probabilistic Inference with Algebraic Constraints: Theoretical Limits and Practical Approximations

A Compositional Atlas of Tractable Circuit Operations for Probabilistic Inference

A Unified Approach to Count-Based Weakly Supervised Learning