ICLR 2026 Past Generative modelsTheory
ICLR 2026 2nd Workshop on Deep Generative Model in Machine Learning: Theory, Principle and Efficacy
ICLR 2026 DeLTa Workshop
- Submission deadline
- Feb 9, 2026, 12:59 UTC imported from OpenReview — check the website for extensions
- Submission portal
- OpenReview
- Notes
- Auto-imported from the OpenReview venue record on 2026-06-10 — please verify and enrich (topics are keyword-guessed).
Accepted papers (133)
Fetched from OpenReview (v2) on 2026-06-10.
-
$\mathbf{R^3}$-Adapter: Progressive Residual Refinement and Representational Alignment for Personalized Image Generation
-
$W_K, W_V$ is Probably All You Need: On the Necessity of the Query, Key, and Value Weight Triplet in Self-Attention Transformers
-
A Complete Decomposition of Stochastic Differential Equations
-
A Diffusive Classification Loss for Learning Energy-based Generative Models
-
A Geometric Perspective on Recursive Synthetic Training
-
A Graph-Theoretical View of Space Folding via the Motzkin–Straus Framework
-
A Unified Density Operator View of Flow Control and Merging
-
Adapting Noise to Data by Quantile Learning
-
AlphaQ: Calibration-Free Bit Allocation for Mixture-of-Experts Quantization
-
An Efficient Test-Time Scaling Approach for Image Generation
-
An Equivariance Toolbox for Learning Dynamics
-
ANCRe: Adaptive Neural Connection Reassignment for Efficient Depth Scaling
-
AnimalBooth: Multimodal Feature Enhancement for Animal Subject Personalization
-
AReUReDi: Annealed Rectified Updates for Refining Discrete Flows with Multi-Objective Guidance
-
Attention Projection Mixing with Exogenous Anchors
-
Avoid What You Know: Divergent Trajectory Balance for GFlowNets
-
B-DENSE: Branching For Dense Ensemble Network Supervision Effeciency
-
Balancing Symmetry and Efficiency in Graph Flow Matching
-
BézierFlow: Learning Bézier Stochastic Interpolant Schedulers for Few-Step Generation
-
BlockGen: Flexible Blockwise Sequence Modeling with Hybrid Samplers
-
BSTabDiff: Block-Subunit Diffusion Priors for High-Dimensional Tabular Data Generation
-
CATS: Inference-aligned SFT for Diffusion LLMs via Context-sensitivity Aware Trajectory Sampling
-
CupOFMoCA: Coupled Objective-Guided Discrete Flows for Molecular Conjugate Assembly
-
Curriculum Sampling: A Two-Phase Curriculum for Efficient Training of Flow Matching
-
Data-Aware Random Feature Kernel for Transformers
-
Decoding Large Language Diffusion Models with Foreseeing Movement
-
Decoupled Diffusion Solver for Inverse Problems on Function Spaces
-
DELTA: Robustly Training Diffusion Models with Weak Annotations
-
Demystifying Transition Matching: When and Why It Can Beat Flow Matching
-
Designing Continuous Conditioning for GANs from WAE Latent Structure
-
Dichotomous Diffusion Policy Optimization
-
Diffusion Models with Double Guidance
-
Diffusion Policy Optimization without Drifting Apart
-
Diffusion Schrödinger Bridge Matching: When Resampling Fails
-
DiffusionShield: A Watermarking Approach to Safeguarding Video Integrity Against Stable Diffusion
-
Dimension-Independent Convergence of Underdamped Langevin Monte Carlo in KL Divergence
-
Discrete Adjoint Schrödinger Bridge Sampler
-
Discrete Bridges for Mutual Information Estimation
-
Discrete Diffusion Samplers and Bridges: Off-Policy Algorithms and Applications in Latent Spaces
-
Discrete Meanflow Training Curriuculum
-
Discriminative Multimodal Preference Models as Guidance for Personalized Image Generation
-
Dynamic Mixture-of-Experts for Visual Autoregressive Model
-
Efficient Tail-Aware Generative Optimization via Flow Model Fine-Tuning
-
Elucidating Guidance in Variance Exploding Diffusion Models: Fast Convergence and Better Diversity
-
Energy-Weighted Flow Matching: Unlocking Continuous Normalizing Flows for Efficient and Scalable Boltzmann Sampling
-
Escaping Model Collapse via Synthetic Data Verification: Near-term Improvements and Long-term Convergence
-
Evaluating the Role of Great Pre-trained Diffusion Models in Few-shot Phase: Warm-up and Acceleration
-
Expert-Data Alignment Governs Generation Quality in Decentralized Diffusion Models
-
Exposing Diversity Bias in Deep Generative Models: Statistical Origins and Correction of Diversity Error
-
Flow Matching based Conditional Independence Tests and Causal Structure Learning
-
Flow Matching in the Low-Noise Regime: Pathologies and a Contrastive Remedy
-
FMMI: Flow Matching Mutual Information Estimation
-
From Compression to Expression: A Layerwise Analysis of In-Context Learning
-
Generative Hints
-
Generative Model via Quantile Assignment
-
Gradual Fine-Tuning for Flow Matching Models
-
Grokking of Diffusion Models: Case Study on Modular Addition
-
GUIDE: Guided Initialization and Distillation of Embeddings
-
Heterogeneous Low-Bandwidth Pre-Training of LLMs
-
Higher-order grammar representations for molecular generation and learning
-
Information-Geometric Optimal Control for Diffusion Models: Unified Framework via Fisher-Rao Geodesics
-
Informative Data Reweighting for Image Classification
-
Inverse Entropic Optimal Transport Solves Semi-supervised Learning via Data Likelihood Maximization
-
Inverse-distilled Diffusion Language Models
-
Latent Process Generator Matching
-
Learning Generation Orders for Masked Discrete Diffusion Models via Variational Inference
-
Learning Unmasking Policies for Diffusion Language Models
-
Log-density Hessian estimation without the curse of dimensionality via denoising score matching
-
Low-Pass Flow Matching
-
Manifold Generalization Provably Proceeds Memorization in Diffusion Models
-
Maximum Entropy under Carre du Champ Constraints
-
Minimal-Action Discrete Schrödinger Bridge Matching for Peptide Sequence Design
-
MixFlow: Mixed Source Distributions Improve Rectified Flows
-
On Closed-Form Couplings
-
On the "Induction Bias" in Sequence Models
-
On the Lipschitz Regularity of Optimal Discriminators
-
On the Memorization of Consistency Distillation for Diffusion Models
-
On the Use of Schrödinger Bridges for Tabular Data Generation
-
One LR Doesn’t Fit All: Heavy-Tail Guided Layerwise Learning Rates for LLMs
-
One-Step Residual Shifting Diffusion for Image Super-Resolution via Distillation
-
Optimal Learning-Rate Schedules under Functional Scaling Laws: Power Decay and Warmup-Stable-Decay
-
Overclocking Electrostatic Generative Models
-
Paired Wasserstein Autoencoders for Conditional Sampling
-
PairFlow: Closed-Form Source-Target Coupling for Few-Step Generation in Discrete Flow Models
-
Path Invariance and the Robustness of Flow Matching: Beyond Architectural and Data Perturbations
-
pCoMole: Pareto-Constrained Molecule Editing with Discrete Flows
-
Performance Limits of Score-Based Generative Models via Stochastic Thermodynamics
-
Permutation-Symmetrized Diffusion for Unconditional Molecular Generation
-
Pre-training Large Language Models with Dynamic Precision: Low-Cost Computation with High-Fidelity Performance
-
Principled Randomized Exploration of Gradient Subspaces for Efficient LLM Training
-
Provable Benefits of RLVR over SFT for Reasoning Models: Learning to Backtrack Efficiently
-
Query Lower Bounds for Diffusion Sampling
-
Rejection Mixing: Fast Semantic Propagation of Mask Tokens for Efficient DLLM Inference
-
Rethinking Reparameterization of Stochastic Processes in Generative Modeling
-
Reward-Guided Discrete Diffusion via Clean-Sample Markov Chain
-
Rex: A Family of Reversible Exponential (Stochastic) Runge-Kutta Solvers
-
RFG: Test-Time Scaling for Diffusion Large Language Model Reasoning with Reward-Free Guidance
-
Robust Graph Diffusion Model
-
Robust Stochastic Gradient Posterior Sampling with Lattice Based Discretisation
-
Scalable Sampling via Generalized Fixed-Point Diffusion Matching
-
Schrödinger bridge problem via empirical risk minimization
-
Score-Guided Proximal Projection: A Unified Geometric Framework for Rectified Flow Editing
-
Search or Accelerate: Confidence-Switched Position Beam Search for Diffusion Language Models
-
SHAPE: SCHEDULE HESSIAN ADAPTIVE PARAMETER ESTIMATION FOR SMOOTHER DIFFUSION OPTIMIZATION
-
SingLoRA: Low Rank Adaptation Using a Single Matrix
-
Skip To The Good Part: Representation Structure & Inference-Time Layer Skipping in Diffusion vs Autoregressive LLM
-
Sliding Critical Band in RoPE-based Length Extrapolation
-
Spectral Condition for $\mu$P under Width–Depth Scaling
-
Steering diffusion models with quadratic rewards: a fine-grained analysis
-
Stein Diffusion Guidance: Training-Free Posterior Correction for Sampling Beyond High-Density Regions
-
Stochastic Few-step Models
-
Strong Reward Only: Pareto-Guided Multi-Reward Optimization
-
Structured image representation learning for flow-matching models
-
Support-Proximity Augmented Diffusion Estimation for Offline Black-Box Optimization
-
Synergistic Intra- and Cross-Layer Regularization Losses for MoE Expert Specialization
-
SYNTHONY: A Stress-Aware, Intent-Conditioned Agent for Deep Tabular Generative Models Selection
-
TD3B: Transition-Directed Discrete Diffusion for Allosteric Binder Generation
-
Time Dependent Loss Reweighting for Flow Matching and Diffusion Models is Theoretically Justified
-
Time-Correlated Video Bridge Matching
-
Tokenize, Diffuse, Decode: A Generative Approach to Neighborhood Discovery on Graphs
-
Training Flow Matching: The Role of Weighting and Parameterization
-
Training-Free Length Discovery for Diffusion Language Model Infilling
-
Understanding Deterministic Diffusion through Reverse Transition Kernels
-
Unleashing the Potential of Diffusion Models for End-to-End Autonomous Driving
-
Unlocking the Duality between Flow and Field Matching
-
Video Unlearning via Low-Rank Refusal Vector
-
What Flow-Matching Brings To TD-Learning
-
What Lies Beneath the Curve? Scaling Laws in the Presence of Exact Posteriors
-
When Does Sparsity Mitigate the Curse of Depth in LLMs
-
When Does Stein Beat Antithetic Sampling? Distribution Complexity in Discrete Gradient Estimation
-
Why Diffusion Language Models Struggle with Truly Parallel (Non-Autoregressive) Decoding?
-
WiSP-OSch: Solver Within-Step Parallelism and Order Scheduling for Diffusion Sampling
-
Zero-Flow Encoders