深度学习 · generative models and autoencoders论文趋势 2023–2026

主要结论

• 深度学习 · generative models and autoencoders论文从 2023 年的 1,216 篇增长到 2026 年的 1,430 篇，三年间扩大到约 1.2 倍。
• NeurIPS 是发深度学习 · generative models and autoencoders论文最多的会议，2023–2026 年共 1,714 篇，占四个会议总量的 40%。
• 交叉方向里涨得最快的是图像生成：同时属于两个主题的论文从 2023 年的 35 篇增至 2026 年的 150 篇。
• 在「深度学习」的 11 个主题里，它的 2026 年论文数排第 2；同组增长最快的扩散模型是 4.1 倍，深度学习 · generative models and autoencoders是 1.2 倍。

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

• 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

全部深度学习 · generative models and autoencoders论文

共 349 篇，第 7 / 7 页

301. NeurIPS 2026Tessellations of Semi-Discrete Flow Matching
302. NeurIPS 2026The $1/\mathcal{W}$ Law: Context Length is the Dominant Energy Lever in LLM Inference Fleets
303. NeurIPS 2026The balance between feature learning and collapse in generative dynamical systems
304. NeurIPS 2026The Benefits of Temporal Correlations: SGD Efficiently Learns k-Juntas from Random Walks
305. NeurIPS 2026The Butterfly Effect in Reasoning: Branch-Structured Distributional Memory for Stochastic Agents
306. NeurIPS 2026The Expressivity Boundary of Probabilistic Circuits: A Comparison with Large Language Models
307. NeurIPS 2026The Fault in Our Metrics: Revisiting Generative Model Evaluation with Chamfer Distance
308. NeurIPS 2026The Rate-Distortion-Polysemanticity Tradeoff in SAEs
309. NeurIPS 2026The Weight Gram Matrix Captures Sequential Feature Linearization in Deep Networks
310. NeurIPS 2026Towards Unified Memory Adaptation for LLM Agents: Textual, Latent, and Parametric Pathways
311. NeurIPS 2026TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation
312. NeurIPS 2026Training-Free Generative Sampling via Moment-Matched Score Smoothing
313. NeurIPS 2026Transporting Quantiles to Codebook: Scalable Vector Quantization without Codebook Collapse
314. NeurIPS 2026Tree-Sliced Orlicz Integral Probability Metric
315. NeurIPS 2026Trie-Aware Transformers for Generative Recommendation
316. NeurIPS 2026Trivialized Generative Models on Lie Groups
317. NeurIPS 2026Tunable Latent Generative Priors for Compressed Sensing and Inverse Problems
318. NeurIPS 2026Twisted Schrödinger Bridge Matching
319. NeurIPS 2026Two Speeds of Learning: A Representation-Readout Decomposition of Grokking and Double Descent
320. NeurIPS 2026UGGRH: Unsupervised Generative Completion and Graph-attention Refinement for Incomplete Cross-modal Hashing
321. NeurIPS 2026Understanding diffusion models requires rethinking (again) generalization
322. NeurIPS 2026Understanding the Challenges in Iterative Generative Optimization with LLMs
323. NeurIPS 2026Understanding the Interplay between Memorization and Learning in Large Language Models
324. NeurIPS 2026Unified Generative-Predictive Modeling for 4D Scene Understanding
325. NeurIPS 2026Uniform Diffusion Models revisited: Leave-One-Out Denoiser and Absorbing State Reformulation
326. NeurIPS 2026Uniform Stability and Generalization Error of GD and SGD on Fixed-Point Parameters
327. NeurIPS 2026UniVer: A Unified Perspective for Multi-step and Multi-draft Speculative Decoding
328. NeurIPS 2026Unleashing the Power of Intrinsic-Entropy-Driven Exploration for Off-Policy Generative RL
329. NeurIPS 2026Unlocking the Duality between Flow and Field Matching
330. NeurIPS 2026Variable-Length Generative Protein Design via Generalized Poisson Flow
331. NeurIPS 2026Vermeer: Autoregressive generative modeling of microscopy predicts protein localization
332. NeurIPS 2026VGB for Masked Diffusion Model: Efficient Test-time Scaling for Reward Satisfaction and Sample Editing
333. NeurIPS 2026Wavelet Flow Matching for Multi-Scale Physics Emulation
334. NeurIPS 2026What Can Labels Alone Audit and Edit in Frozen Representations?
335. NeurIPS 2026What Cohort INRs Encode, and Where to Freeze Them
336. NeurIPS 2026What Does a Sparse Autoencoder Feature Do? A Weight-Based Account
337. NeurIPS 2026What Drives Compositional Generalization? The Importance of Continuous Training Objectives in Visual Generative Models
338. NeurIPS 2026What Makes a Good Path? Factoring Manifold Support and Path Geometry
339. NeurIPS 2026What Matters for Diffusion-Friendly Latent Manifold? Prior-Aligned Autoencoders for Latent Diffusion
340. NeurIPS 2026When and How to Canonize: a Generalization Perspective
341. NeurIPS 2026When and Why Adversarial Training Improves PINNs: A Neural Tangent Kernel Perspective
342. NeurIPS 2026When and Why SignSGD Outperforms SGD: A Theoretical Study Based on $\ell_1$-norm Lower Bounds
343. NeurIPS 2026When Are Compositional Problems Learnable from Verifiable Rewards?
344. NeurIPS 2026When Edge Independence Fails: Joint Graph Diffusion with Latent Sociability Priors
345. NeurIPS 2026When Less is More: The LLM Scaling Paradox in Context Compression
346. NeurIPS 2026When Metropolis and Hastings Meet Bradley and Terry: Exact MCMC From Preference Voting
347. NeurIPS 2026Why Cross-Skeleton Retargeting Is Non-Identifiable: Structural Limits of Generative Motion Models
348. NeurIPS 2026Why Pass@k Optimization Can Degrade Pass@1: Prompt Interference in LLM Post-Training
349. NeurIPS 2026WorldComposer: Generative High-Fidelity Simulation with Digital Cousins for Generalizable Robot Learning and Evaluation

常见问题

2023–2026 年四大会议一共有多少篇深度学习 · generative models and autoencoders论文？

共 4,314 篇，按年份分别是 2023 年 1,216 篇、2024 年 768 篇、2025 年 900 篇、2026 年 1,430 篇。

NeurIPS 2026 有多少篇深度学习 · generative models and autoencoders论文？

NeurIPS 2026 有 349 篇深度学习 · generative models and autoencoders论文，上一届是 240 篇。

ICLR 2026 有多少篇深度学习 · generative models and autoencoders论文？

ICLR 2026 有 303 篇深度学习 · generative models and autoencoders论文，上一届是 275 篇。

ICML 2026 有多少篇深度学习 · generative models and autoencoders论文？

ICML 2026 有 604 篇深度学习 · generative models and autoencoders论文，上一届是 275 篇。

CVPR 2026 有多少篇深度学习 · generative models and autoencoders论文？

CVPR 2026 有 174 篇深度学习 · generative models and autoencoders论文，上一届是 110 篇。

哪个会议的深度学习 · generative models and autoencoders论文最多？

NeurIPS，2023–2026 年共 1,714 篇。