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

1,350 papers, page 1 of 27

1. ICLR 2026A Derandomization Framework for Structure Discovery: Applications in Neural Networks and Beyond
2. ICLR 2026A Law of Data Reconstruction for Random Features (And Beyond)
3. ICLR 2026A Near-Optimal Best-of-Both-Worlds Algorithm for Federated Bandits
4. ICLR 2026A New Approach to Controlling Linear Dynamical Systems
5. ICLR 2026A New Initialization to Control Gradients in Sinusoidal Neural Networks
6. ICLR 2026A Recovery Guarantee for Sparse Neural Networks
7. ICLR 2026A Representer Theorem for Hawkes Processes via Penalized Least Squares Minimization
8. ICLR 2026A Sharp KL-Convergence Analysis for Diffusion Models under Minimal Assumptions
9. ICLR 2026A Statistical Theory of Overfitting for Imbalanced Classification
10. ICLR 2026A Theoretical Analysis of Mamba’s Training Dynamics: Filtering Relevant Features for Generalization in State Space Models
11. ICLR 2026A Unifying View of Coverage in Linear Off-policy Evaluation
12. ICLR 2026Achieving Approximate Symmetry Is Exponentially Easier than Exact Symmetry
13. ICLR 2026Action Chunking and Data Augmentation Yield Exponential Improvements in Behavior Cloning for Continuous Spaces
14. ICLR 2026Active Learning for Decision Trees with Provable Guarantees
15. ICLR 2026Adversarially Pretrained Transformers may be Universally Robust In-Context Learners
16. ICLR 2026Almost Bayesian: Dynamics of SGD Through Singular Learning Theory
17. ICLR 2026An efficient, provably optimal, practical algorithm for the 0-1 loss linear classification problem
18. ICLR 2026An Improved Model-free Decision-estimation Coefficient with Applications in Adversarial MDPs
19. ICLR 2026Automata Learning and Identification of the Support of Language Models
20. ICLR 2026Bandit Learning in Matching Markets Robust to Adversarial Corruptions
21. ICLR 2026Barriers for Learning in an Evolving World: Mathematical Understanding of Loss of Plasticity
22. ICLR 2026Benefits and Limitations of Communication in Multi-Agent Reasoning
23. ICLR 2026Best-of-Majority: Minimax-Optimal Strategy for Pass@k Inference Scaling
24. ICLR 2026Best-of-N through the Smoothing Lens: KL Divergence and Regret Analysis
25. ICLR 2026Best-of-three-worlds Analysis for Dueling Bandits with Borda Winner
26. ICLR 2026Better Bounds for the Distributed Experts Problem
27. ICLR 2026Beyond Spectra: Eigenvector Overlaps in Loss Geometry
28. ICLR 2026Block-Sample MAC-Bayes Generalization Bounds
29. ICLR 2026Bridging Kolmogorov Complexity and Deep Learning: Asymptotically Optimal Description Length Objectives for Transformers
30. ICLR 2026Characterizing the Discrete Geometry of ReLU Networks
31. ICLR 2026Closed-form $\ell_r$ norm scaling with data for overparameterized linear regression and diagonal linear networks under $\ell_p$ bias
32. ICLR 2026Computational Bottlenecks for Denoising Diffusions
33. ICLR 2026Computing Equilibrium beyond Unilateral Deviation
34. ICLR 2026Dataset Distillation for Memorized Data: Soft Labels can Leak Held-Out Teacher Knowledge
35. ICLR 2026Deep-ICE: The first globally optimal algorithm for empirical risk minimization of two-layer maxout and ReLU networks
36. ICLR 2026Deterministic Bounds and Random Estimates of Metric Tensors on Neuromanifolds
37. ICLR 2026Differentially Private Equilibrium Finding in Polymatrix Games
38. ICLR 2026Differentially Private Two-Stage Gradient Descent for Instrumental Variable Regression
39. ICLR 2026Diffusion Language Models are Provably Optimal Parallel Samplers
40. ICLR 2026Dimension-Free Decision Calibration for Nonlinear Loss Functions
41. ICLR 2026Dimension-Free Minimax Rates for Learning Pairwise Interactions in Attention-Style Models
42. ICLR 2026Directional Convergence, Benign Overfitting of Gradient Descent in leaky ReLU two-layer Neural Networks
43. ICLR 2026Distributed Algorithms for Euclidean Clustering
44. ICLR 2026Diverse Dictionary Learning
45. ICLR 2026Diversified Multinomial Logit Contextual Bandits
46. ICLR 2026Does the Data Processing Inequality Reflect Practice? On the Utility of Low-Level Tasks
47. ICLR 2026Does Weak-to-strong Generalization Happen under Spurious Correlations?
48. ICLR 2026Dynamical properties of dense associative memory
49. ICLR 2026Efficient Adversarial Attacks on High-dimensional Offline Bandits
50. ICLR 2026Efficient Testing for Correlation Clustering: Improved Algorithms and Optimal Bounds

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.