Accepted Papers

Below is the list of accepted papers:

title decision Poster Section
Self-Speculative Decoding in Any-Order and Any-Subset Autoregressive Models Accept (Oral) Poster Section 1
Learning to Iteratively Improve 3D Representation with 2D Generative Models Accept Poster Section 1
Zero-Variance Gradients for Variational Autoencoders Accept Poster Section 1
Towards Practical Multi-label Causal Discovery in High-Dimensional Event Sequences via One-Shot Graph Aggregation Accept Poster Section 1
CoVAE: Consistency Training of Variational Autoencoders Accept Poster Section 1
Shaping Inductive Bias in Diffusion Models through Frequency-Based Noise Control Accept Poster Section 1
Inference-time Scaling of Diffusion Models through Classical Search Accept (Oral) Poster Section 1
SLayR: Scene Layout Generation with Rectified Flow Accept Poster Section 1
Rethinking Direct Preference Optimization in Diffusion Models Accept Poster Section 1
Insertion Language Models: Sequence Generation with Arbitrary-Position Insertions Accept Poster Section 1
Probabilistic Soundness Guarantees in LLM Reasoning Chains Accept Poster Section 1
Transformers as Unrolled Inference in Probabilistic Laplacian Eigenmaps Accept Poster Section 1
Inference and Generating Method for Extremely Sparse Networks Accept Poster Section 1
Scalable Bayesian Monte Carlo: fast uncertainty estimation beyond deep ensembles Accept Poster Section 1
MMG: Mutual Information Estimation via the MMSE Gap in Diffusion Accept Poster Section 1
Posterior Inference in Latent Space for Scalable Constrained Black-box Optimization Accept (Oral) Poster Section 1
The Unwinnable Arms Race of AI Image Detection Accept Poster Section 1
Improving Iterative Gaussian Processes via Warm Starting Sequential Posteriors Accept Poster Section 1
FlowBack-Adjoint: Energy-Guided Conditional Flow-Matching for Protein Side-Chain Generation Accept Poster Section 1
State-Space Architectures for Scalable Diffusion-based 3D Molecule Generation Accept Poster Section 1
Partial Resolution: Optimal Multi-Draft Speculative Sampling in Near-Linear Time Accept Poster Section 1
STED and Consistency Scoring: A Framework for Evaluating LLM Structured Output Reliability Accept Poster Section 1
Enhancing Diffusion Model Guidance through Calibration and Regularization Accept Poster Section 1
An Information-Theoretic Discrete Poisson Diffusion Framework Accept Poster Section 1
Variational Deep Learning via Implicit Regularization Accept Poster Section 1
Semantic Volume: Quantifying and Detecting both External and Internal Uncertainty in LLMs Accept Poster Section 1
From Entropy Rate to Redundancy: Information Dynamics in Large Language Models Accept Poster Section 1
A Multi-Method Interpretability Framework for Probing Cognitive Processing in Deep Neural Networks across Vision and Biomedical Domains Accept Poster Section 1
moPPIt-v3: Motif-Specific Peptides Generated via Multi-Objective-Guided Discrete Flow Matching Accept Poster Section 1
Information-Guided Diffusion Sampling for Dataset Distillation Accept Poster Section 1
Token-Level Guided Discrete Diffusion for Membrane Protein Design Accept Poster Section 1
A Probabilistic Approach to Pose Synchronization for Multi-Reference Alignment with applications to wireless communication systems Accept Poster Section 1
Graph Random Features for Scalable Gaussian Processes Accept Poster Section 1
GFlowNets for Learning Better Drug-Drug Interaction Representations Accept Poster Section 1
BP-Seg: A graphical model approach to unsupervised and non-contiguous text segmentation using belief propagation Accept Poster Section 1
Failure Prediction Is a Better Performance Proxy for Early-Exit Networks Than Calibration Accept Poster Section 1
Value Matching: Scalable and Gradient-Free Reward-Guided Flow Adaptation Accept Poster Section 1
Bayes-PD: Exploring a Sequence to Binding Bayesian Neural Network model trained on Phage Display data Accept Poster Section 1
Simple, Fast and Efficient Injective Manifold Density Estimation with Random Projections Accept Poster Section 1
Blind Inverse Problem Solving Made Easy by Text-to-Image Latent Diffusion Accept Poster Section 1
DDS-E-Sim: A Transformer-based Probabilistic Generative Framework for Simulating Error-Prone DNA Sequences for DNA Data Storage Accept Poster Section 1
Score-informed Neural Operator for Enhancing Ordering-based Causal Discovery Accept Poster Section 1
Hold That Exit: Near Optimal Early-Exit Inference via Recall Accept Poster Section 1
Scaffold Diffusion: Sparse Multi-Category Voxel Structure Generation with Discrete Diffusion Accept Poster Section 1
VarDiU: A Variational Diffusive Upper Bound for One-Step Diffusion Distillation Accept Poster Section 1
Steering Pretrained Drafters during Speculative Decoding Accept Poster Section 1
Tokenized Neural Fields: Structured Representations of Continuous Signals Accept Poster Section 1
TwinTURBO: Semi-Supervised Fine-Tuning of Foundation Models via Mutual Information Decompositions for Downstream Task and Latent Spaces Accept Poster Section 1
Multi-Objective Nanobody Design via Masked Discrete Diffusion with Simplex Refinement Accept Poster Section 1
Weighted Conditional Flow Matching Accept Poster Section 1
A Connection Between Score Matching and Local Intrinsic Dimension Accept Poster Section 2
Multimodal Bayesian Network for Robust Assessment of Casualty in Autonomous Triage Accept Poster Section 2
Selective Underfitting in Diffusion Models Accept Poster Section 2
Reconsidering Noise for Denoising Diffusion Probabilistic Models Accept Poster Section 2
Effective Diffusion-free Score Matching for Exact Conditional Sampling Accept Poster Section 2
Learning Boltzmann Generators via Constrained Mass Transport Accept (Oral) Poster Section 2
Personalized English Amharic Medical Image Caption and Speech Generation for Visually Impaired Patients Using Vision Transformer Fused with LLM Accept Poster Section 2
Trust Region Constrained Measure Transport in Path Space for Stochastic Optimal Control and Inference Accept Poster Section 2
Is Sequence Information All You Need for Bayesian Optimization of Antibodies? Accept Poster Section 2
Can We Estimate The Entropy Of Arbitrary Distributions Known Up To A Normalization Constant? Accept Poster Section 2
Probabilistic Image Generation with LLM Priors via Structured Rectified Flow Accept Poster Section 2
Are We Really Learning the Score Function? Reinterpreting Diffusion Models Through Wasserstein Gradient Flow Matching Accept Poster Section 2
On Fitting Flow Models with Large Sinkhorn Couplings Accept Poster Section 2
Improving Generation Quality of Long-Tailed Diffusion via Disentangled Latent Representations Accept Poster Section 2
Generalization of Diffusion Models Arises from a Regularized Representation Space Accept Poster Section 2
3-Model Speculative Decoding Accept Poster Section 2
Cross-Lingual Multimodal Retrieval-Augmented Generation for Open Question Answering in Tamil and Yoruba Accept Poster Section 2
Beyond Linear Diffusions: Improved Representations for Rare Conditional Generative Modeling Accept Poster Section 2
Temporal Alignment Guidance: On-manifold Sampling in Diffusion Models Accept Poster Section 2
ISUM: Inverse Problem Solver via Unbalanced Optimal Transport Map Accept Poster Section 2
BioBO: Biology-informed Bayesian Optimization for Perturbation Design Accept Poster Section 2
Myosotis: structured computation for attention like layer Accept Poster Section 2
Generative Actor-Critic Accept Poster Section 2
Learning Velocity Prior-Guided Hamiltonian-Jacobi Flows with Unbalanced Optimal Transport Accept Poster Section 2
Constrained Molecular Generation via Sequential Flow Model Fine-Tuning Accept Poster Section 2
Accelerating Diffusion Models in Offline RL via Reward-Aware Consistency Trajectory Distillation Accept Poster Section 2
Entropy-Guided Sampling of Flat Modes in Discrete Spaces Accept Poster Section 2
Slithering through Gaps: Capturing Discrete Isolated Modes via Logistic Bridging Accept Poster Section 2
Leveraging Probabilistic Modeling for Robust End-to-End Autonomous Driving across Domains Accept Poster Section 2
A Theory of Multi-Agent Generative Flow Networks Accept Poster Section 2
Robust Multi-task Modeling for Bayesian Optimization via In-Context Learning Accept Poster Section 2
When Do LLMs Improve Bayesian Optimization? A Systematic Comparison Across Molecular and Protein Design Accept Poster Section 2
Divergence Minimization Preference Optimization for Diffusion Model Alignment Accept Poster Section 2
Ambient Diffusion Omni Accept (Oral) Poster Section 2
GenUQ: Predictive Uncertainty Estimates via Generative Hyper-Networks Accept Poster Section 2
Inception Inference: Nested Probabilistic Reasoning over Story Graphs from Text Accept Poster Section 2
Continuous-Token Diffusion for Speaker-Referenced TTS in Multimodal LLMs Accept Poster Section 2
IAGA: Identity-Aware Gaussian Approximation for Efficient 3D Molecular Generation Accept Poster Section 2
Score-based Idempotent Distillation of Diffusion Models Accept Poster Section 2
SpectFlow: Long-term forecasting using flow matching with 89k parameters Accept Poster Section 2
Neural Universal Scene Descriptors Accept Poster Section 2
ImmUQBench: A Benchmark on Uncertainty Quantification of Protein Immunogenicity Prediction Accept Poster Section 2
SpecAttn - Speculating Sparse Attention Accept Poster Section 2
Efficient Flow Matching using Latent Variables Accept Poster Section 2
Any-Order Flexible Length Masked Diffusion Accept Poster Section 2
PolUQBench: A Benchmark Study on Uncertainty Quantification of Polymer Property Prediction Accept Poster Section 2
Contrastive MIM: A Contrastive Mutual Information Framework for Unified Generative and Discriminative Representation Learning Accept Poster Section 2
Adaptive Frontier Exploration on Graphs with Applications to Network-Based Disease Testing Accept Poster Section 2
Using maximal information auxiliary variables to improve synthetic data generation based on TabPFN foundation models Accept Poster Section 2
Probabilistic Variational Contrastive Learning Accept Poster Section 2
GNN-Guided Block Selection in Gibbs MCMC Accept Poster Section 2
CoPE: A Lightweight Complex Positional Encoding Accept Poster Section 2
An Optimal Algorithm for Marginalization in Bayesian Networks Accept Poster Section 2
Semantic Probabilistic Control of Language Models Accept Poster Section 2
Foundations of Top-$k$ Decoding for Language Models Accept Poster Section 2
Improved Sampling from Masked Diffusion Models with Position Contrastive Guidance Accept Poster Section 2
When rule learning breaks: Diffusion Fails to Learn Parity of Many Bits Accept (Oral) Poster Section 2
DenseMixer: Improving MoE Post-Training with Precise Router Gradient Accept Poster Section 2
Multi-scale Autoregressive Models are Laplacian, Discrete, and Latent Diffusion Models In Disguise Accept Poster Section 2
Where the Score Lives: A Wavelet View of Diffusion Accept Poster Section 2
Value Gradient Guidance for Flow Matching Alignment Accept Poster Section 2
Entangled Schrödinger Bridge Matching Accept Poster Section 2
Entropy Is Not Enough: Uncertainty Quantification for LLMs fails under Aleatoric Uncertainty Accept Poster Section 2
ScooBDoob: Schrödinger Bridge with Doob’s h-Transform for Molecular Dynamics Accept Poster Section 2
Diffusion Beats Autoregressive in Data-Constrained Settings Accept Poster Section 2