Theory · learning theory papers, 2023–2026

Key findings

• Theory · learning theory went from 86 papers in 2023 to 482 in 2026, roughly 5.6x in three years.
• NeurIPS publishes the most Theory · learning theory work: 540 papers from 2023 to 2026, 40% of the four-venue total.
• The fastest-growing overlap is with Social aspects · accountability transparency and interpretability: papers tagged with both rose from 0 in 2023 to 33 in 2026.
• Among the 5 topics under Theory and Statistics, it ranks #2 by 2026 paper count; the fastest grower in the group, Theory · optimization, is up 4.7x versus 5.6x here.

Representative papers

Spread across years and venues so every edition shows up

• NeurIPS 2026A dynamical systems theory of reward-modulated learning in linear recurrent networks
• ICML 2026A Fine-Grained Understanding of Uniform Convergence for Halfspaces
• ICLR 2026A Law of Data Reconstruction for Random Features (And Beyond)
• NeurIPS 2025A Geometric Analysis of PCA
• ICML 2025A Generalization Theory for Zero-Shot Prediction
• ICLR 2025A Solvable Attention for Neural Scaling Laws
• CVPR 2025Dynamic Stereotype Theory Induced Micro-expression Recognition with Oriented Deformation
• NeurIPS 20244+3 Phases of Compute-Optimal Neural Scaling Laws
• ICML 2024A Statistical Theory of Regularization-Based Continual Learning
• ICLR 2024A Sublinear Adversarial Training Algorithm
• CVPR 2024Accurate Training Data for Occupancy Map Prediction in Automated Driving Using Evidence Theory
• NeurIPS 2023$\varepsilon$-fractional core stability in Hedonic Games.
• ICML 2023Robust Collaborative Learning with Linear Gradient Overhead
• ICLR 2023Bispectral Neural Networks
• NeurIPS 2026A Mechanistic Investigation of Theory of Mind in a Large Language Model
• ICML 2026A Random Matrix Theory of Masked Self-Supervised Learning
• ICLR 2026A New Approach to Controlling Linear Dynamical Systems
• NeurIPS 2025Approximation theory for 1-Lipschitz ResNets
• ICML 2025A Two-Stage Learning-to-Defer Approach for Multi-Task Learning
• ICLR 2025A Theoretical Analysis of Self-Supervised Learning for Vision Transformers

All Theory · learning theory papers

132 papers, page 2 of 3

51. ICML 2025Identifying Metric Structures of Deep Latent Variable Models
52. ICML 2025Implicit Regularization for Tubal Tensor Factorizations via Gradient Descent
53. ICML 2025Improved Learning via k-DTW: A Novel Dissimilarity Measure for Curves
54. ICML 2025Incorporating Arbitrary Matrix Group Equivariance into KANs
55. ICML 2025Laplace Transform Based Low-Complexity Learning of Continuous Markov Semigroups
56. ICML 2025Learning Configurations for Data-Driven Multi-Objective Optimization
57. ICML 2025Learning In-context $n$-grams with Transformers: Sub-$n$-grams Are Near-Stationary Points
58. ICML 2025Learning Parametric Distributions from Samples and Preferences
59. ICML 2025Learning to Steer Learners in Games
60. ICML 2025Learning Utilities from Demonstrations in Markov Decision Processes
61. ICML 2025Learning with Exact Invariances in Polynomial Time
62. ICML 2025Learning-Augmented Hierarchical Clustering
63. ICML 2025Limitations of measure-first protocols in quantum machine learning
64. ICML 2025LoRA-One: One-Step Full Gradient Could Suffice for Fine-Tuning Large Language Models, Provably and Efficiently
65. ICML 2025Mastering Multiple-Expert Routing: Realizable $H$-Consistency and Strong Guarantees for Learning to Defer
66. ICML 2025Matrix Completion with Incomplete Side Information via Orthogonal Complement Projection
67. ICML 2025Meta Optimality for Demographic Parity Constrained Regression via Post-Processing
68. ICML 2025Metastable Dynamics of Chain-of-Thought Reasoning: Provable Benefits of Search, RL and Distillation
69. ICML 2025Mixture of Experts Provably Detect and Learn the Latent Cluster Structure in Gradient-Based Learning
70. ICML 2025Model Uncertainty Quantification by Conformal Prediction in Continual Learning
71. ICML 2025Models of Heavy-Tailed Mechanistic Universality
72. ICML 2025Modular Duality in Deep Learning
73. ICML 2025Nearly Optimal Sample Complexity for Learning with Label Proportions
74. ICML 2025No Free Lunch from Random Feature Ensembles: Scaling Laws and Near-Optimality Conditions
75. ICML 2025Non-Asymptotic Length Generalization
76. ICML 2025Nonconvex Theory of $M$-estimators with Decomposable Regularizers
77. ICML 2025Nonlinear transformers can perform inference-time feature learning
78. ICML 2025On Learning Parallel Pancakes with Mostly Uniform Weights
79. ICML 2025On the Convergence of Continuous Single-timescale Actor-critic
80. ICML 2025On the Emergence of Position Bias in Transformers
81. ICML 2025On the Generalization Ability of Next-Token-Prediction Pretraining
82. ICML 2025On the Learnability of Distribution Classes with Adaptive Adversaries
83. ICML 2025On the Role of Label Noise in the Feature Learning Process
84. ICML 2025On the Similarities of Embeddings in Contrastive Learning
85. ICML 2025On the Statistical Mechanisms of Distributional Compositional Generalization
86. ICML 2025On the Training Convergence of Transformers for In-Context Classification of Gaussian Mixtures
87. ICML 2025PAC Learning with Improvements
88. ICML 2025PDUDT: Provable Decentralized Unlearning under Dynamic Topologies
89. ICML 2025Policy-Regret Minimization in Markov Games with Function Approximation
90. ICML 2025Position: Algebra Unveils Deep Learning - An Invitation to Neuroalgebraic Geometry
91. ICML 2025Position: An Empirically Grounded Identifiability Theory Will Accelerate Self Supervised Learning Research
92. ICML 2025Probably Approximately Global Robustness Certification
93. ICML 2025Projection Optimization: A General Framework for Multi-Objective and Multi-Group RLHF
94. ICML 2025Provable Benefits of Unsupervised Pre-training and Transfer Learning via Single-Index Models
95. ICML 2025Provable Efficiency of Guidance in Diffusion Models for General Data Distribution
96. ICML 2025Provably Efficient RL for Linear MDPs under Instantaneous Safety Constraints in Non-Convex Feature Spaces
97. ICML 2025Rapid Overfitting of Multi-Pass SGD in Stochastic Convex Optimization
98. ICML 2025Refined generalization analysis of the Deep Ritz Method and Physics-Informed Neural Networks
99. ICML 2025Representation Preserving Multiclass Agnostic to Realizable Reduction
100. ICML 2025Representative Language Generation

FAQ

How many Theory · learning theory papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

1,350 in total: 86 in 2023, 393 in 2024, 389 in 2025, 482 in 2026.

How many Theory · learning theory papers are at NeurIPS 2026?

NeurIPS 2026 has 112 Theory · learning theory papers, versus 106 the year before.

How many Theory · learning theory papers are at ICLR 2026?

ICLR 2026 has 200 Theory · learning theory papers, versus 147 the year before.

How many Theory · learning theory papers are at ICML 2026?

ICML 2026 has 170 Theory · learning theory papers, versus 132 the year before.

How many Theory · learning theory papers are at CVPR 2025?

CVPR 2025 has 4 Theory · learning theory papers, versus 1 the year before.

Which conference publishes the most Theory · learning theory papers?

NeurIPS, with 540 papers from 2023 to 2026.