通用机器学习 · unsupervised and semisupervised learning论文趋势 2023–2026

主要结论

• 通用机器学习 · unsupervised and semisupervised learning论文从 2023 年的 649 篇增长到 2026 年的 799 篇，三年间扩大到约 1.2 倍。
• ICML 是发通用机器学习 · unsupervised and semisupervised learning论文最多的会议，2023–2026 年共 859 篇，占四个会议总量的 32%。
• 交叉方向里涨得最快的是深度学习 · self supervised learning：同时属于两个主题的论文从 2023 年的 23 篇增至 2026 年的 86 篇。
• 在「通用机器学习」的 6 个主题里，它的 2026 年论文数排第 1；同组增长最快的通用机器学习 · evaluation是 4.3 倍，通用机器学习 · unsupervised and semisupervised learning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 general kernel framework for non-CND distance measures using $|\mathcal{D}|$-dimensional sparse landmark embeddings
• ICML 2026$\mu$pscaling small models: Principled warm starts and hyperparameter transfer
• ICLR 2026A Generalized Geometric Theoretical Framework of Centroid Discriminant Analysis for Linear Classification of Multi-dimensional Data
• CVPR 2026COG Confidence-aware Optimal Geometric Correspondence for Unsupervised Single-reference Novel Object
• NeurIPS 2025$O(\sqrt{T})$ Static Regret and Instance Dependent Constraint Violation for Constrained Online Convex Optimization
• ICML 2025A Square Peg in a Square Hole: Meta-Expert for Long-Tailed Semi-Supervised Learning
• ICLR 2025A General Framework for Producing Interpretable Semantic Text Embeddings
• CVPR 2025A General Adaptive Dual-level Weighting Mechanism for Remote Sensing Pansharpening
• NeurIPS 2024GlotCC: An Open Broad-Coverage CommonCrawl Corpus and Pipeline for Minority Languages
• ICML 2024CKGConv: General Graph Convolution with Continuous Kernels
• ICLR 2024A Graph is Worth 1-bit Spikes: When Graph Contrastive Learning Meets Spiking Neural Networks
• CVPR 2024Dual Prototype Attention for Unsupervised Video Object Segmentation
• NeurIPS 2023$H$-Consistency Bounds: Characterization and Extensions
• ICML 2023A Mathematical Model for Curriculum Learning for Parities
• ICLR 2023A General Framework For Proving The Equivariant Strong Lottery Ticket Hypothesis
• CVPR 2023A General Regret Bound of Preconditioned Gradient Method for DNN Training
• NeurIPS 2026Attention-based PCA
• ICML 2026A Control-Theoretic View of Mamba on Stability and Robustness
• ICLR 2026A Single Architecture for Representing Invariance Under Any Space Group
• CVPR 2026CROWn A Unified Framework for Anti-Aliased Downsampling and Phase-Calibrated Fusion

全部通用机器学习 · unsupervised and semisupervised learning论文

共 798 篇，第 8 / 16 页

351. ICML 2026Distilling Linearized Behavior into Non-linear Fine-Tuning for Effective Task Arithmetic
352. ICML 2026Divisiveness-Consistent Label Distribution Learning
353. ICML 2026DMCO: Budget-Aware Co-Optimization of Data Cleaning and AutoML
354. ICML 2026Do LLMs Signal When They’re Right? Evidence from Neuron Agreement
355. ICML 2026Doc-to-LoRA: Learning to Instantly Internalize Contexts
356. ICML 2026Domain Adaptation with Adaptive $f$-Divergence: Tighter Variational Representation and Generalization Bounds
357. ICML 2026DomED: Redesigning Ensemble Distillation for Domain Generalization
358. ICML 2026DSGym: A Standardized and Holistic Framework for Evaluating and Training Data Science Agents
359. ICML 2026DuetServe: Harmonizing Prefill and Decode for LLM Serving via Adaptive GPU Multiplexing
360. ICML 2026Dynamic High-Dimensional Facility Location with Low Recourse
361. ICML 2026ECSEL: Explainable Classification via Signomial Equation Learning
362. ICML 2026Efficient and Minimax Optimal In-context Nonparametric Regression with Transformers
363. ICML 2026Efficient Synthetic Network Generation via Latent Embedding Reconstruction
364. ICML 2026Efficiently Solving Discounted MDPs via Predictions with Unknown Prediction Errors
365. ICML 2026Ekka: Automated Diagnosis of Silent Errors in LLM Inference
366. ICML 2026Embedding Hybrid Systems into Continuous Latent Vector Fields
367. ICML 2026Emergent Visual Representations through Unsupervised Spiking Networks with Synaptic Pruning
368. ICML 2026Enhanced Multi-Instance Partial Label Learning via Average Gradient Outer Product
369. ICML 2026Enhancing Cross-subject Emotion Recognition via Heterogeneous Distribution Augmentation and Collaborative Learning
370. ICML 2026Equivalence of Context and Parameter Updates in Modern Transformer Blocks
371. ICML 2026Equivariant Neural Networks for General Linear Symmetries on Lie Algebras
372. ICML 2026Escaping the Subspace Trap: The Role of Optimizer Geometry in Model Width Expansion
373. ICML 2026Estimating Correlation Clustering Cost in Node-Arrival Stream
374. ICML 2026Exact Functional ANOVA Decomposition for Categorical Inputs Models
375. ICML 2026Expand Neurons, Not Parameters
376. ICML 2026Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift
377. ICML 2026Explaining Concept Shift with Interpretable Feature Attribution
378. ICML 2026Explicitly Modeling Censoring Produces Superior Survival Predictors
379. ICML 2026Fantastic Reasoning Behaviors and Where to Find Them: Unsupervised Discovery of the Reasoning Process
380. ICML 2026FaPS: A General and Fast Training Method for Diffusion Models
381. ICML 2026Fast k-means Seeding Under The Manifold Hypothesis
382. ICML 2026Fast kernel methods: Sobolev, physics-informed, and additive models
383. ICML 2026Faster Activation Functions at the Edge for Post-Training Speedups
384. ICML 2026Federated Data and Feature Selection by Generalized CUR Decomposition
385. ICML 2026Federated Learning with Unlabeled Clients: Personalization Can Happen in Low Dimensions
386. ICML 2026Fine-Tune Once, Reuse Across Models: Bayesian Task-Update Factors and Approximations
387. ICML 2026Fix the Loss, Not the Radius: Rethinking the Adversarial Perturbation of Sharpness-Aware Minimization
388. ICML 2026Fixed Budget is No Harder Than Fixed Confidence in Best-Arm Identification up to Logarithmic Factors
389. ICML 2026FluxNet: Learning Capacity-Constrained Local Transport Operators for Conservative and Bounded PDE Surrogates
390. ICML 2026FOCUS: DLLMs Know How to Tame Their Compute Bound
391. ICML 2026Formalizing and Falsifying Causal Pathways of Rare Events
392. ICML 2026Foundations of Equivariant Deep Learning: Unifying Graph and Sheaf Neural Networks
393. ICML 2026From Directions to Regions: Decomposing Activations in Language Models via Local Geometry
394. ICML 2026From Generalist to Specialist Representation
395. ICML 2026From Memorization to Parameter Interference: How Overtraining Experts Harms Model Merging
396. ICML 2026Fully Dynamic Coreset Spectral Clustering
397. ICML 2026FUSE: Ensembling Verifiers with Zero Labeled Data
398. ICML 2026General Analysis of LMO-based Optimizers: Beyond Bounded Variance
399. ICML 2026General and Efficient Steering of Unconditional Diffusion Models
400. ICML 2026General Quantification of Covariate and Concept Shifts

常见问题

2023–2026 年四大会议一共有多少篇通用机器学习 · unsupervised and semisupervised learning论文？

共 2,683 篇，按年份分别是 2023 年 649 篇、2024 年 553 篇、2025 年 682 篇、2026 年 799 篇。

NeurIPS 2026 有多少篇通用机器学习 · unsupervised and semisupervised learning论文？

NeurIPS 2026 有 194 篇通用机器学习 · unsupervised and semisupervised learning论文，上一届是 203 篇。

ICLR 2026 有多少篇通用机器学习 · unsupervised and semisupervised learning论文？

ICLR 2026 有 187 篇通用机器学习 · unsupervised and semisupervised learning论文，上一届是 174 篇。

ICML 2026 有多少篇通用机器学习 · unsupervised and semisupervised learning论文？

ICML 2026 有 348 篇通用机器学习 · unsupervised and semisupervised learning论文，上一届是 239 篇。

CVPR 2026 有多少篇通用机器学习 · unsupervised and semisupervised learning论文？

CVPR 2026 有 70 篇通用机器学习 · unsupervised and semisupervised learning论文，上一届是 66 篇。

哪个会议的通用机器学习 · unsupervised and semisupervised learning论文最多？

ICML，2023–2026 年共 859 篇。