Deep learning · generative models and autoencoders papers, 2023–2026

Key findings

• Deep learning · generative models and autoencoders went from 1,216 papers in 2023 to 1,430 in 2026, roughly 1.2x in three years.
• NeurIPS publishes the most Deep learning · generative models and autoencoders work: 1,714 papers from 2023 to 2026, 40% of the four-venue total.
• The fastest-growing overlap is with Image Generation: papers tagged with both rose from 35 in 2023 to 150 in 2026.
• Among the 11 topics under Deep Learning, it ranks #2 by 2026 paper count; the fastest grower in the group, Diffusion Models, is up 4.1x versus 1.2x here.

Representative papers

Spread across years and venues so every edition shows up

• NeurIPS 2026A Critical $\beta$-Scale for Posterior Collapse in Dirichlet $\beta$-VAEs
• ICML 2026(1D) Ordered Tokens Enable Efficient Test-Time Search
• ICLR 2026``Noisier'’ Noise Contrastive Estimation is (Almost) Maximum Likelihood
• CVPR 2026Anchoring and Rescaling Attention for Semantically Coherent Inbetweening
• NeurIPS 2025Agnostic Learning under Targeted Poisoning: Optimal Rates and the Role of Randomness
• ICML 2025A Likelihood Based Approach to Distribution Regression Using Conditional Deep Generative Models
• ICLR 2025A Watermark for Order-Agnostic Language Models
• CVPR 2025DUNE: Distilling a Universal Encoder from Heterogeneous 2D and 3D Teachers
• NeurIPS 2024[Re] GNNInterpreter: A probabilistic generative model-level explanation for Graph Neural Networks
• ICML 2024A Fixed-Point Approach for Causal Generative Modeling
• ICLR 2024$t^3$-Variational Autoencoder: Learning Heavy-tailed Data with Student's t and Power Divergence
• CVPR 2024Decompose-and-Compose: A Compositional Approach to Mitigating Spurious Correlation
• NeurIPS 2023(S)GD over Diagonal Linear Networks: Implicit bias, Large Stepsizes and Edge of Stability
• ICML 2023A Kernelized Stein Discrepancy for Biological Sequences
• ICLR 2023Arbitrary Virtual Try-on Network: Characteristics Representation and Trade-off between Body and Clothing
• CVPR 20233D GAN Inversion With Facial Symmetry Prior
• NeurIPS 2026Action-On-Item Preference Flow: A Shared Event Schema for Predictive and Generative Personalization
• ICML 2026$f$-Trajectory Balance: A Loss Family for Tuning GFlowNets, Generative Models, and LLMs with Off- and On-Policy Data
• ICLR 2026Analyzing and Evaluating Unbiased Language Model Watermark
• CVPR 2026Bidirectional Normalizing Flow From Data to Noise and Back

All Deep learning · generative models and autoencoders papers

4,314 papers, page 22 of 87

1051. ICML 2026Var-JEPA: A Variational Formulation of the Joint-Embedding Predictive Architecture – Bridging Predictive and Generative Self-Supervised Learning
1052. ICML 2026Variable-Length Tokenization via Learnable Global Merging for Diffusion Transformers
1053. ICML 2026Variance Driven Exploration: A Provable and Efficient Methodology for Pure Exploration in Highly Stochastic Environments
1054. ICML 2026Variational Bayesian Flow Network for Graph Generation
1055. ICML 2026Variational Entropic Optimal Transport
1056. ICML 2026Variational Flow Maps: Make Some Noise for One-Step Conditional Generation
1057. ICML 2026VecDesigner: Exploring Visual Guidance and Structural Consistency for Semantic Typography
1058. ICML 2026VectorWorld: Efficient Streaming World Model via Diffusion Flow on Vector Graphs
1059. ICML 2026VELR: Efficient Video Reward Feedback via Ensemble Latent Reward Models
1060. ICML 2026VideoFlexTok: Flexible-Length Coarse-to-Fine Video Tokenization
1061. ICML 2026VideoGPA: Distilling Geometry Priors for 3D-Consistent Video Generation
1062. ICML 2026VideoMAETok: Boosting Video Diffusion Models via Masked Autoencoders as Tokenizers
1063. ICML 2026Visual Implicit Autoregressive Modeling
1064. ICML 2026ViTok-v2: Scaling Native Resolution Autoencoders to 5 Billion Parameters
1065. ICML 2026Watch Your Step: Information Injection in Diffusion Models via Shadow Timestep Embedding
1066. ICML 2026Weak Diffusion Priors Can Still Achieve Strong Inverse-Problem Performance
1067. ICML 2026When Diffusion Language Models Hesitate: Detecting and Correcting Visual Hallucinations via Confidence Fluctuation
1068. ICML 2026When Do Diffusion Models Learn to Generate Multiple Objects?
1069. ICML 2026When Model Merging Breaks Routing: Training-Free Calibration for MoE
1070. ICML 2026When RL Meets Adaptive Speculative Training: A Unified Training-Serving System
1071. ICML 2026When Sample Selection Bias Precipitates Model Collapse
1072. ICML 2026Where Concept Erasure Should Occur: Concept–Layer Alignment in Text-to-Video Diffusion Models
1073. ICML 2026Where Detectors Fail: Probing Generative Space for Generalizable AI-Generated Image Detection
1074. ICML 2026Where Rectified Flows Leak: Characterising Membership Signals Along the Interpolation Path
1075. ICML 2026WhisperSplat: Lossless Steganography in 3D Gaussian Splatting
1076. ICML 2026Who Said Neural Networks Aren't Linear?
1077. ICML 2026Why DDIM Hallucinates More Than DDPM: A Theoretical Analysis of Reverse Dynamics
1078. ICML 2026Why Deep Jacobian Spectra Separate: Depth-Induced Scaling and Singular-Vector Alignment
1079. ICML 2026XSpecMesh: Quality-Preserving Auto-Regressive Mesh Generation Acceleration via Multi-Head Speculative Decoding
1080. ICML 2026You Can Learn Tokenization End-to-End with Reinforcement Learning
1081. ICML 2026You Don’t Need All That Attention: Surgical Memorization Mitigation in Text-to-Image Diffusion Models
1082. NeurIPS 2026$\mathcal{P}$Torch: Narrowing the Gap Between Projection and Gradient-Based Learning
1083. NeurIPS 2026A Critical $\beta$-Scale for Posterior Collapse in Dirichlet $\beta$-VAEs
1084. NeurIPS 2026A Locally Tokenized Generative Model for Robust Time-Series Watermarking
1085. NeurIPS 2026A Mean-Field Framework for Inference-Time Distributional Control of Diffusion Models
1086. NeurIPS 2026A simple model of co-emergence of grid and place fields
1087. NeurIPS 2026A Unifying Perspective on Language Model Interpretability
1088. NeurIPS 2026ABC: Any-Subset Autoregression via Non-Markovian Diffusion Bridges in Continuous Time and Space
1089. NeurIPS 2026Action-On-Item Preference Flow: A Shared Event Schema for Predictive and Generative Personalization
1090. NeurIPS 2026Active Learning for Conditional Generative Compressed Sensing
1091. NeurIPS 2026Active Learning of Conditional Generative Models via the Transport Neural Tangent Kernel
1092. NeurIPS 2026Aegis: Generative Gradient Masking for Privacy-Preserving Medical Federated Learning
1093. NeurIPS 2026AegisFlow: Training-free Non-myopic Path-safe Guided Flow Matching
1094. NeurIPS 2026AesGI-Bench: Benchmarking and Evaluating the Aesthetic Quality of AI-Generated Images via Large Multimodal Models
1095. NeurIPS 2026AI-Assisted Classification under Correlation Neglect and Trust
1096. NeurIPS 2026Align $\&$ Invert: Solving Inverse Problems with Diffusion and Flow-based Models via Representation Alignment
1097. NeurIPS 2026Align as You Couple: Learning Spatial Resolved Inference from H&E Images with Mollified Flow Matching
1098. NeurIPS 2026Aligning Flow Map Policies with Optimal $Q$-Guidance
1099. NeurIPS 2026All-Addition Spiking Diffusion Models with Attention Enhancement
1100. NeurIPS 2026AMGenC: Generating Charge Balanced Amorphous Materials

FAQ

How many Deep learning · generative models and autoencoders papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

4,314 in total: 1,216 in 2023, 768 in 2024, 900 in 2025, 1,430 in 2026.

How many Deep learning · generative models and autoencoders papers are at NeurIPS 2026?

NeurIPS 2026 has 349 Deep learning · generative models and autoencoders papers, versus 240 the year before.

How many Deep learning · generative models and autoencoders papers are at ICLR 2026?

ICLR 2026 has 303 Deep learning · generative models and autoencoders papers, versus 275 the year before.

How many Deep learning · generative models and autoencoders papers are at ICML 2026?

ICML 2026 has 604 Deep learning · generative models and autoencoders papers, versus 275 the year before.

How many Deep learning · generative models and autoencoders papers are at CVPR 2026?

CVPR 2026 has 174 Deep learning · generative models and autoencoders papers, versus 110 the year before.

Which conference publishes the most Deep learning · generative models and autoencoders papers?

NeurIPS, with 1,714 papers from 2023 to 2026.