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

328 papers, page 1 of 7

1. ICML 2026A Decision-Theoretic View of Test-Time Training: When, How Far, and Which Directions to Adapt
2. ICML 2026A Fine-Grained Understanding of Uniform Convergence for Halfspaces
3. ICML 2026A General Framework for Fair and Robust Regression
4. ICML 2026A Random Matrix Perspective on the Consistency of Diffusion Models
5. ICML 2026A Random Matrix Theory of Masked Self-Supervised Learning
6. ICML 2026A Refined Generalization Analysis for Extreme Multi-class Supervised Contrastive Representation Learning
7. ICML 2026A Risk Decomposition Framework for Pre-hoc Fine-tuning Prediction
8. ICML 2026A Solvable High-Dimensional Model Where Nonlinear Autoencoders Learn Structure Invisible to PCA While Test Loss Misaligns With Generalization
9. ICML 2026A Task-centric Theory for Iterative Self-Improvement with Easy-to-Hard Curricula
10. ICML 2026A Theoretical Framework for Statistical Evaluability of Generative Models
11. ICML 2026A Theory of How Pretraining Shapes Inductive Bias in Fine-Tuning
12. ICML 2026A theory of learning data statistics in diffusion models, from easy to hard
13. ICML 2026Active Learning with Low-Rank Structure for Data Selection
14. ICML 2026Adaptive Preconditioners Trigger Loss Spikes in Adam
15. ICML 2026Adversarially Robust Approximate Furthest Neighbor
16. ICML 2026AI4SLT: Empirical Processes in Lean 4 for Formal Statistical Learning Theory
17. ICML 2026Alignment-Sensitive Minimax Rates for Spectral Algorithms with Learned Kernels
18. ICML 2026All ERMs Can Fail in Stochastic Convex Optimization Lower Bounds in Linear Dimension
19. ICML 2026An Algebraic View of the Expressivity of Recurrent Language Models
20. ICML 2026An analytic theory of convolutional neural network inverse problems solvers
21. ICML 2026An Efficient Joint Learning Approach for Item Response Theory
22. ICML 2026Approximate Equivariance via Projection-Based Regularisation
23. ICML 2026Approximation Error Upper and Lower Bounds for Hölder Class with Transformers
24. ICML 2026Approximation Theory for Lipschitz Continuous Transformers
25. ICML 2026Asymptotic Theory of Iterated Empirical Risk Minimization, with Applications to Active Learning
26. ICML 2026Asymptotically Fast Clebsch-Gordan Tensor Products with Vector Spherical Harmonics
27. ICML 2026Benign Overfitting in Adversarial Training for Vision Transformers
28. ICML 2026Beyond Binary: Continuous State Optimization with Graph-Structured Objectives
29. ICML 2026Black-Box Assisted Regression: Phase Transitions and Minimax Optimality
30. ICML 2026Breaking the Reversal Curse in Autoregressive Language Models via Identity Bridge
31. ICML 2026Certifying Capabilities from Finite Tests: When Is It Possible?
32. ICML 2026Characterization of Gaussian Universality Breakdown in High-Dimensional Empirical Risk Minimization
33. ICML 2026Characterizing the Effect of Noise in Language Generation in the Limit
34. ICML 2026Combinatorial Sparse PCA Beyond the Spiked Identity Model
35. ICML 2026Consistency Deep Equilibrium Models
36. ICML 2026Continuous Variable Hamiltonian Learning at Heisenberg Limit via Displacement-Random Unitary Transformation
37. ICML 2026Convex Distance Operator Transport: A Convex and Geometry-Preserving Formulation
38. ICML 2026Correcting Overparameterization Effects in Fair Empirical Risk Minimization
39. ICML 2026Coupled Training with Privileged Information and Unlabeled Data
40. ICML 2026Coverage Improvement and Fast Convergence of On-policy Preference Learning
41. ICML 2026Dataset Distillation Efficiently Encodes Low-Dimensional Representations from Gradient-Based Learning of Non-Linear Tasks
42. ICML 2026Decision Tree Learning on Product Spaces
43. ICML 2026Decomposing Out-of-Distribution Error in Conditional Flow Matching via Wasserstein Geometry
44. ICML 2026Deep Learning of Compositional Targets with Hierarchical Spectral Methods
45. ICML 2026Deep Networks Learn Deep Hierarchical Models
46. ICML 2026Deep Neural Network Regression with Functional Covariates
47. ICML 2026Desirable Effort Fairness and Optimality Trade-offs in Strategic Learning
48. ICML 2026Diffusion Models Are Statistically Optimal for Learning Low-Dimensional Multi-Modal Distributions
49. ICML 2026DPO Unchained: Your Training Algorithm is Secretly Disentangled in Human Choice Theory (and Its Loss' Convexity is Dispensable)
50. ICML 2026Effective Model Pruning : Measuring the Redundancy of Model Components

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.