Workshop Sat, Dec 6, 2025 • 8:00 AM – 5:00 PM PST Upper Level Ballroom 20C

Structured Probabilistic Inference and Generative Modeling

Yuanqi Du · Dinghuai Zhang · Jiajun He · Heli Ben-Hamu · Francisco Vargas · Yunan Yang · Animashree Anandkumar · Arnaud Doucet · José Miguel Hernández-Lobato

Project Page [ OpenReview]

Video

Chat is not available.

Schedule

Timezone: America/Los_Angeles

8:30 AM

Opening remark

8:40 AM

Invited talk (Jiaxin Shi)

9:10 AM

Sponsor talk (Basis research institute)

9:40 AM

Invited talk (Chelsea Finn, Yoonho Lee)

9:55 AM

Sponsor talk (Genesis Molecular AI)

10:10 AM

Coffee break

10:30 AM

Invited talk (Andrew Gordon Wilson)

11:00 AM

Contributed talk #1

11:30 AM

Poster session #1

12:30 PM

Panel: Is probabilistic inference still relevant in the era of foundation models?

1:30 PM

Invited talk (Arash Vahdat)

2:00 PM

Invited talk (Ruiqi Gao)

2:30 PM

Coffee break

2:40 PM

Invited talk (Giulia Luise)

3:10 PM

Contributed talk #2

3:40 PM

Closing remark

3:45 PM

Poster session #2

Ambient Diffusion Omni

Giannis Daras · Adrian Rodriguez-Munoz · Adam Klivans · Antonio Torralba · Constantinos Daskalakis

Link

SLayR: Scene Layout Generation with Rectified Flow

Cameron Braunstein · Hevra Petekkaya · Jan Eric Lenssen · Mariya Toneva · Eddy Ilg

Link

Enhancing Diffusion Model Guidance through Calibration and Regularization

Seyed Alireza Javid · Amirhossein Bagheri · Nuria Prelcic

Link

A Multi-Method Interpretability Framework for Probing Cognitive Processing in Deep Neural Networks across Vision and Biomedical Domains

Harshini Suresha · Kavitha H

Link

DDS-E-Sim: A Transformer-based Probabilistic Generative Framework for Simulating Error-Prone DNA Sequences for DNA Data Storage

Mst. Naznin · Swarup Sidhartho Mondol · Adnan Faruq · Debashmita Saha · Ahmed Rumi · A. B. M. Alim Al Islam

Link

ISUM: Inverse Problem Solver via Unbalanced Optimal Transport Map

Donggyu Lee · Taekyung Lee · Jaewoong Choi · Myungjoo Kang

Link

Scaffold Diffusion: Sparse Multi-Category Voxel Structure Generation with Discrete Diffusion

Justin Jung

Link

Divergence Minimization Preference Optimization for Diffusion Model Alignment

Binxu Li · Minkai Xu · Meihua Dang · Stefano Ermon

Link

moPPIt-v3: Motif-Specific Peptides Generated via Multi-Objective-Guided Discrete Flow Matching

Tong Chen · Zachary Quinn · Yinuo Zhang · Pranam Chatterjee

Link

Simple, Fast and Efficient Injective Manifold Density Estimation with Random Projections

Ahmad Ayaz Amin

Link

Using maximal information auxiliary variables to improve synthetic data generation based on TabPFN foundation models

Elias Chaibub Neto

Link

Scalable Bayesian Monte Carlo: fast uncertainty estimation beyond deep ensembles

Xinzhu Liang · Joseph Lukens · Brian Kirby · Thomas Searles · Sanjaya Lohani · Xin Qiu · Kody Law

Link

STED and Consistency Scoring: A Framework for Evaluating LLM Structured Output Reliability

Guanghui Wang · Jinze Yu · Xing Zhang · Dayuan jiang · Yin Song · Peiyang He · Xuefeng Liu · Tomal Deb

Link

Shaping Inductive Bias in Diffusion Models through Frequency-Based Noise Control

Thomas Jiralerspong · Berton Earnshaw · Jason Hartford · Yoshua Bengio · Luca Scimeca

Link

An Information-Theoretic Discrete Poisson Diffusion Framework

Sagnik Bhattacharya · Abhiram Gorle · Ahsan Bilal · Amit Kumar Singh Yadav · Connor Ding · Tsachy Weissman

Link

Effective Diffusion-free Score Matching for Exact Conditional Sampling

Thomas Wedenig · Robert Peharz

Link

Partial Resolution: Optimal Multi-Draft Speculative Sampling in Near-Linear Time

Rahul Thomas · Arka Pal

Link

Transformers as Unrolled Inference in Probabilistic Laplacian Eigenmaps

Aditya Ravuri · Neil Lawrence

Link

Information-Guided Diffusion Sampling for Dataset Distillation

Linfeng Ye · Shayan Mohajer Hamidi · Guang Li · Takahiro Ogawa · Miki Haseyama · Konstantinos N Plataniotis

Link

BP-Seg: A graphical model approach to unsupervised and non-contiguous text segmentation using belief propagation

Fengyi Li · Kayhan Behdin · Xiaofeng Wang · Ercan Yildiz · Natesh Pillai · Zhipeng Wang

Link

Value Matching: Scalable and Gradient-Free Reward-Guided Flow Adaptation

Cristian Jensen · Luca Schaufelberger · Riccardo De Santi · Kjell Jorner · Andreas Krause

Link

Diffusion Beats Autoregressive in Data-Constrained Settings

Mihir Prabhudesai · Mengning Wu · Amir Zadeh · Katerina Fragkiadaki · Deepak Pathak

Link

Generative Actor-Critic

Aoyang Qin · Wei Wang · Deqian Kong · Ying Nian Wu · Song-Chun Zhu · Sirui Xie

Link

Probabilistic Variational Contrastive Learning

Minoh Jeong · Seonho Kim · Alfred Hero

Link

Inference-time Scaling of Diffusion Models through Classical Search

Xiangcheng Zhang · Haowei Lin · Haotian Ye · James Zou · Jianzhu Ma · Yitao Liang · Yilun Du

Link

MMG: Mutual Information Estimation via the MMSE Gap in Diffusion

Longxuan Yu · Xing Shi · Xianghao Kong · Tong Jia · Greg Ver Steeg

Link

Can We Estimate The Entropy Of Arbitrary Distributions Known Up To A Normalization Constant?

Safa Messaoud · Skander Charni · Elaa Bouazza · Ali Pourghasemi Fatideh · Halima Bensmail

Link

GNN-Guided Block Selection in Gibbs MCMC

Benjamin Dayan · Bhaskar Mishra · Justin Svegliato · Stuart J Russell

Link

Entropy-Guided Sampling of Flat Modes in Discrete Spaces

Pinaki Mohanty · Riddhiman Bhattacharya · Ruqi Zhang

Link

Inception Inference: Nested Probabilistic Reasoning over Story Graphs from Text

Gokul Srinath Seetha Ram

Link

Bayes-PD: Exploring a Sequence to Binding Bayesian Neural Network model trained on Phage Display data

AMIAUD--PLACHY Ilann · Oliver Bent · Sebastien Boyer

Link

Adaptive Frontier Exploration on Graphs with Applications to Network-Based Disease Testing

XianJun, Davin Choo · Yuqi Pan · Tonghan Wang · Milind Tambe · Alastair van Heerden · Cheryl Johnson

Link

Where the Score Lives: A Wavelet View of Diffusion

Emma Finn · Binxu Wang · Andy Keller · Demba Ba

Link

VarDiU: A Variational Diffusive Upper Bound for One-Step Diffusion Distillation

Leyang Wang · Mingtian Zhang · Zijing Ou · David Barber

Link

Trust Region Constrained Measure Transport in Path Space for Stochastic Optimal Control and Inference

Denis Blessing · Julius Berner · Lorenz Richter · Carles Domingo i Enrich · Yuanqi Du · Arash Vahdat · Gerhard Neumann

Link

Probabilistic Image Generation with LLM Priors via Structured Rectified Flow

Mykola Vysotskyi · Zahar Kohut · Anna-Alina Bondarets · Taras Rumezhak

Link

A Theory of Multi-Agent Generative Flow Networks

Leo Brunswic · Haozhi Wang · Shuang Luo · Jianye Hao · Amir Rasouli · Yinchuan Li

Link

Improved Sampling from Masked Diffusion Models with Position Contrastive Guidance

Dhruvesh Patel · Tahira Naseem · Gaurav Pandey · Md Arafat Sultan · Andrew McCallum · Ramón Astudillo

Link

Hold That Exit: Near Optimal Early-Exit Inference via Recall

Yuanyuan Yang · Ruimin Zhang · Jamie Morgenstern · Haifeng Xu

Link

Beyond Linear Diffusions: Improved Representations for Rare Conditional Generative Modeling

Kulunu Dharmakeerthi · Yousef El-Laham · Henry Wong · Vamsi Potluru · Changhong He · Taosong He

Link

Any-Order Flexible Length Masked Diffusion

Jaeyeon Kim · Lee Kit · Carles Domingo i Enrich · Yilun Du · Sham Kakade · Timothy Ngotiaoco · Sitan Chen · Michael Albergo

Link

Probabilistic Soundness Guarantees in LLM Reasoning Chains

Weiqiu You · Anton Xue · Shreya Havaldar · Delip Rao · Helen Jin · Chris Callison-Burch · Eric Wong

Link

ImmUQBench: A Benchmark on Uncertainty Quantification of Protein Immunogenicity Prediction

Alif Bin Abdul Qayyum · Amir Hossein Rahmati · Xiaoning Qian · Byung-Jun Yoon

Link

The Unwinnable Arms Race of AI Image Detection

Till Aczel · Lorenzo Vettor · Andreas Plesner · Roger Wattenhofer

Link

Variational Deep Learning via Implicit Regularization

Jonathan Wenger · Beau Coker · Juraj Marusic · John Cunningham

Link

Constrained Molecular Generation via Sequential Flow Model Fine-Tuning

Sven Gutjahr · Riccardo De Santi · Luca Schaufelberger · Kjell Jorner · Andreas Krause

Link

Accelerating Diffusion Models in Offline RL via Reward-Aware Consistency Trajectory Distillation

Xintong Duan · Yutong He · Fahim Tajwar · Ruslan Salakhutdinov · Zico Kolter · Jeff Schneider

Link

Zero-Variance Gradients for Variational Autoencoders

Zilei Shao · Anji Liu · Guy Van den Broeck

Link

Token-Level Guided Discrete Diffusion for Membrane Protein Design

Shrey Goel · Peregrine Schray · Yinuo Zhang · Sophia Vincoff · Huong Kratochvil · Pranam Chatterjee

Link

From Entropy Rate to Redundancy: Information Dynamics in Large Language Models

Jessica Liang

Link

Improving Iterative Gaussian Processes via Warm Starting Sequential Posteriors

Alan Dong · Jihao Andreas Lin · José Miguel Hernández-Lobato

Link

Graph Random Features for Scalable Gaussian Processes

Matthew Zhang · Jihao Andreas Lin · Krzysztof M Choromanski · Adrian Weller · Richard Turner · Isaac Reid

Link

Leveraging Probabilistic Modeling for Robust End-to-End Autonomous Driving across Domains

Rajeev Yasarla · Shizhong Han · Hsin-Pai Cheng · Litian Liu · Shweta Mahajan · Apratim Bhattacharyya · Yunxiao Shi · Risheek Garrepalli · Herbert Cai · Fatih Porikli

Link

Posterior Inference in Latent Space for Scalable Constrained Black-box Optimization

Kiyoung Om · Kyuil Sim · Taeyoung Yun · Hyeongyu Kang · Jinkyoo Park

Link

An Optimal Algorithm for Marginalization in Bayesian Networks

Alina Yang · Bhaskar Mishra

Link

Semantic Probabilistic Control of Language Models

Kareem Ahmed · Catarina Belém · Padhraic Smyth · Sameer Singh

Link

Inference and Generating Method for Extremely Sparse Networks

Valentin Kilian · Benjamin Guedj · Francois Caron

Link

Reconsidering Noise for Denoising Diffusion Probabilistic Models

Stephen Liang

Link

Temporal Alignment Guidance: On-manifold Sampling in Diffusion Models

Youngrok Park · Hojung Jung · Sangmin Bae · Se-Young Yun

Link

Cross-Lingual Multimodal Retrieval-Augmented Generation for Open Question Answering in Tamil and Yoruba

Kiran Raja · Mobareji Abejide · Arya Ram · Utkarsh Sharma · Benjamin Liu · Kevin Zhu

Link

Multi-Objective Nanobody Design via Masked Discrete Diffusion with Simplex Refinement

Ruoxi Zhang · Pranam Chatterjee

Link

Continuous-Token Diffusion for Speaker-Referenced TTS in Multimodal LLMs

Xinlu He · Harish Mallidi · JIA-HONG HUANG · Ashwin Bellur · M. Maruf · Chander Chandak · Venkatesh Ravichandran · Swayambhu Ray

Link

Generalization of Diffusion Models Arises from a Regularized Representation Space

Zekai Zhang · Xiang Li · Xiao Li · Molei Tao · Qing Qu

Link

Entangled Schrödinger Bridge Matching

Sophia Tang · Yinuo Zhang · Pranam Chatterjee

Link

Score-informed Neural Operator for Enhancing Ordering-based Causal Discovery

Jiyeon Kang · Songseong Kim · Chanhui Lee · Doyeong Hwang · Joanie Chung · Yunkyung Ko · Sumin Lee · Sungwoong Kim · Sungbin Lim

Link

TwinTURBO: Semi-Supervised Fine-Tuning of Foundation Models via Mutual Information Decompositions for Downstream Task and Latent Spaces

Guillaume Quétant · Pavlo Molchanov · Slava Voloshynovskiy

Link

Blind Inverse Problem Solving Made Easy by Text-to-Image Latent Diffusion

Michail Dontas · Yutong He · Naoki Murata · Yuki Mitsufuji · Zico Kolter · Ruslan Salakhutdinov

Link

Slithering through Gaps: Capturing Discrete Isolated Modes via Logistic Bridging

Pinaki Mohanty · Ruqi Zhang

Link

BioBO: Biology-informed Bayesian Optimization for Perturbation Design

Yanke Li · Tianyu Cui · Tommaso Mansi · Mangal Prakash · Rui Liao

Link

On Fitting Flow Models with Large Sinkhorn Couplings

Alireza Mousavi-Hosseini · Stephen Zhang · Michal Klein · Marco Cuturi

Link

Multimodal Bayesian Network for Robust Assessment of Casualty in Autonomous Triage

Szymon Rusiecki · Cecilia G Morales · Kimberly Elenberg · Leonard Weiss · Artur Dubrawski

Link

SpectFlow: Long-term forecasting using flow matching with 89k parameters

Seyed Mohamad Moghadas · Adrian Munteanu

Link

A Connection Between Score Matching and Local Intrinsic Dimension

Eric Yeats · Aaron Jacobson · Darryl Hannan · Yiran Jia · Tim Doster · Henry Kvinge · Scott Mahan

Link

When rule learning breaks: Diffusion Fails to Learn Parity of Many Bits

Binxu Wang · Emma Finn · Bingbin Liu

Link

GFlowNets for Learning Better Drug-Drug Interaction Representations

Azmine Toushik Wasi

Link

Are We Really Learning the Score Function? Reinterpreting Diffusion Models Through Wasserstein Gradient Flow Matching

An Vuong · Michael McCann · Javier E. Santos · Yen Ting Lin

Link

CoVAE: Consistency Training of Variational Autoencoders

Gianluigi Silvestri · Luca Ambrogioni

Link

PolUQBench: A Benchmark Study on Uncertainty Quantification of Polymer Property Prediction

Alif Bin Abdul Qayyum · Byung-Jun Yoon

Link

Insertion Language Models: Sequence Generation with Arbitrary-Position Insertions

Dhruvesh Patel · Aishwarya Sahoo · Avinash Amballa · Tahira Naseem · Tim G. J. Rudner · Andrew McCallum

Link

GenUQ: Predictive Uncertainty Estimates via Generative Hyper-Networks

Tian Yen · Reese Jones · Ravi Patel

Link

Learning Velocity Prior-Guided Hamiltonian-Jacobi Flows with Unbalanced Optimal Transport

Amy Xiang Wang

Link

Robust Multi-task Modeling for Bayesian Optimization via In-Context Learning

Yucen Li · Samuel Daulton · Samuel Müller · Andrew Wilson · Eytan Bakshy

Link

When Do LLMs Improve Bayesian Optimization? A Systematic Comparison Across Molecular and Protein Design

Mattias Akke · Soojung Yang · Jurģis Ruža · Jinyeop Song · Elton Pan · Rafael Gomez-Bombarelli

Link

Contrastive MIM: A Contrastive Mutual Information Framework for Unified Generative and Discriminative Representation Learning

Micha Livne

Link

Self-Speculative Decoding in Any-Order and Any-Subset Autoregressive Models

Gabe Guo · Stefano Ermon

Link

Failure Prediction Is a Better Performance Proxy for Early-Exit Networks Than Calibration

Piotr Kubaty · Filip Szatkowski · Metod Jazbec · Bartosz Wójcik

Link

Neural Universal Scene Descriptors

Alejandro Escontrela · Shrinu Kushagra · Sjoerd van Steenkiste · Yulia Rubanova · Aleksander Holynski · Kelsey Allen · Kevin Murphy · Thomas Kipf

Link

Multi-scale Autoregressive Models are Laplacian, Discrete, and Latent Diffusion Models In Disguise

Steve Hong · Samuel Belkadi

Link

ScooBDoob: Schrödinger Bridge with Doob’s h-Transform for Molecular Dynamics

Yinuo Zhang · Sophia Tang · Pranam Chatterjee

Link

Score-based Idempotent Distillation of Diffusion Models

Shehtab Zaman · Chengyan Liu · Kenneth Chiu

Link

Improving Generation Quality of Long-Tailed Diffusion via Disentangled Latent Representations

Esther Rodriguez · Monica Welfert · Samuel McDowell · Nathan Stromberg · Julian Camarena · Lalitha Sankar

Link

Personalized English Amharic Medical Image Caption and Speech Generation for Visually Impaired Patients Using Vision Transformer Fused with LLM

Dawit Bogale · Vukosi Marivate · Tesfa Tegegne · Munir Awol

Link

A Probabilistic Approach to Pose Synchronization for Multi-Reference Alignment with applications to wireless communication systems

Rob Romijnders · Gabriele Cesa · Christos Louizos · Kumar Pratik · Arash Behboodi

Link

IAGA: Identity-Aware Gaussian Approximation for Efficient 3D Molecular Generation

Jingxiang Qu · Wenhan Gao · RUICHEN XU · Yi Liu

Link

Efficient Flow Matching using Latent Variables

Anirban Samaddar · Yixuan Sun · Viktor Nilsson · Sandeep Madireddy

Link

DenseMixer: Improving MoE Post-Training with Precise Router Gradient

Feng Yao · Junxia Cui · Ruohan Zhang · Liyuan Liu · Shibo Hao · Li Zhang · Chengyu Dong · Shuohang Wang · yelong shen · Jianfeng Gao · Jingbo Shang

Link

State-Space Architectures for Scalable Diffusion-based 3D Molecule Generation

Adrita Das · Peiran Jiang · Dantong Zhu · Barnabas Poczos · Jose Lugo-Martinez

Link

Entropy Is Not Enough: Uncertainty Quantification for LLMs fails under Aleatoric Uncertainty

Tim Tomov · Dominik Fuchsgruber · Tom Wollschläger · Stephan Günnemann

Link

SpecAttn - Speculating Sparse Attention

Harsh Shah

Link

Selective Underfitting in Diffusion Models

Kiwhan Song · Jaeyeon Kim · Sitan Chen · Yilun Du · Sham Kakade · Vincent Sitzmann

Link

CoPE: A Lightweight Complex Positional Encoding

Avinash Amballa

Link

Semantic Volume: Quantifying and Detecting both External and Internal Uncertainty in LLMs

Xiaomin Li · Zhou Yu · Ziji Zhang · Yingying Zhuang · Swair Shah · Narayanan Sadagopan · Anurag Beniwal

Link

Myosotis: structured computation for attention like layer

Evgenii Egorov · Herbert Cai · Hanno Ackermann · Markus Nagel

Link

Steering Pretrained Drafters during Speculative Decoding

Frédéric Berdoz · Peer Rheinboldt · Roger Wattenhofer

Link

Is Sequence Information All You Need for Bayesian Optimization of Antibodies?

Sebastian Ober · Calvin McCarter · Aniruddh Raghu · Yucen Li · Alan Amin · Andrew Wilson · Hunter Elliott

Link

FlowBack-Adjoint: Energy-Guided Conditional Flow-Matching for Protein Side-Chain Generation

Alexander Berlaga · Michael Jones · Andrew Ferguson

Link

Tokenized Neural Fields: Structured Representations of Continuous Signals

Azmi Haider · Dan Rosenbaum

Link

Value Gradient Guidance for Flow Matching Alignment

Zhen Liu · Tim Xiao · Carles Domingo i Enrich · Weiyang Liu · Dinghuai Zhang

Link

Rethinking Direct Preference Optimization in Diffusion Models

Junyong Kang · Seohyun Lim · Kyungjune Baek · Hyunjung Shim

Link

Learning to Iteratively Improve 3D Representation with 2D Generative Models

Titas Anciukevicius

Link

3-Model Speculative Decoding

Sanghyun Byun · Mohanad Odema · Jung Ick Guack · Baisub Lee · Jacob Song · Woo Seong Chung

Link

Foundations of Top-$k$ Decoding for Language Models

Georgy Noarov · Soham Mallick · Tao Wang · Sunay Joshi · Yan Sun · Yangxinyu Xie · Mengxin Yu · Edgar Dobriban

Link

Weighted Conditional Flow Matching

Sergio Calvo Ordoñez · Matthieu Meunier · Alvaro Cartea · Christoph Reisinger · Yarin Gal · José Miguel Hernández-Lobato

Link

Learning Boltzmann Generators via Constrained Mass Transport

Christopher von Klitzing · Denis Blessing · Henrik Schopmans · Pascal Friederich · Gerhard Neumann

Link

Towards Practical Multi-label Causal Discovery in High-Dimensional Event Sequences via One-Shot Graph Aggregation

Hugo Math · Rainer Lienhart

Link