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

482 papers, page 6 of 10

251. ICML 2026Efficient and Minimax Optimal In-context Nonparametric Regression with Transformers
252. ICML 2026Efficiently Learning Drifting Halfspaces with Massart Noise
253. ICML 2026Envy-Free Allocation of Indivisible Goods via Noisy Queries
254. ICML 2026Equilibrium Pricing in Oligopolistic Data Markets
255. ICML 2026Error Analysis of Discrete Flow with Generator Matching
256. ICML 2026Error Propagation in Dynamic Programming: From Stochastic Control to American Option Pricing
257. ICML 2026Evidential Copula Concept Embedding Models
258. ICML 2026Explaining Data Mixing Scaling Laws
259. ICML 2026Fast Reconstruction of Mixtures of Bernoulli Product Distributions
260. ICML 2026Feature Bagging Provides Stability
261. ICML 2026Flat Minima and Generalization: Insights from Stochastic Convex Optimization
262. ICML 2026Fractional is Better: Learnable Derivative Orders in Neural Operator Learning
263. ICML 2026From Generative to Episodic: Sample-Efficient Replicable Reinforcement Learning
264. ICML 2026Full-Batch Gradient Descent Outperforms One-Pass SGD: Sample Complexity Separation in Single-Index Learning
265. ICML 2026Generalization Bounds for Out-of-distribution Generalization
266. ICML 2026Generalization of Gibbs and Langevin Monte Carlo Algorithms in the Interpolation Regime
267. ICML 2026Gradient Flow Through Diagram Expansions: Learning Regimes and Explicit Solutions
268. ICML 2026Harmful Overfitting in Sobolev Spaces
269. ICML 2026High-Accuracy Sampling for Diffusion Models and Log-Concave Distributions
270. ICML 2026High-Dimensional Learning Dynamics of Quantized Models with Straight-Through Estimator
271. ICML 2026Holonomy Grid Codes for Generalisation Under Directed Actions
272. ICML 2026How Hard Is Science?
273. ICML 2026Improved Distribution Estimation in $\ell_\infty$
274. ICML 2026In-Context Learning Is Provably Bayesian Inference: A Generalization Theory for Meta-Learning
275. ICML 2026Information-Theoretic Generalization Bounds for VAEs: A Role of Encoder and Latent Variable
276. ICML 2026Instance-Dependent Continuous-Time Reinforcement Learning via Maximum Likelihood Estimation
277. ICML 2026Is Spurious Correlation Removal Always Learnable?
278. ICML 2026Joint-Space Empowerment as a Theory of Dexterous Motor Coordination
279. ICML 2026Language Generation in the Limit: Complexity Barriers and Implications for Learning
280. ICML 2026Language Generation with Feedback: Queries and Mistakes
281. ICML 2026Language Generation with Replay: A Learning-Theoretic View of Model Collapse
282. ICML 2026Learning Gaussian Graphical Models from a Glauber Trajectory Without Mixing
283. ICML 2026Learning High-Dimensional Parity Functions with Product Networks using Gradient Descent
284. ICML 2026Learning Junta Distributions, Quantum Junta States, and QAC$^0$ Circuits
285. ICML 2026Learning Partial Concept Classes and Universal Rates Under Massart Noise
286. ICML 2026Learning Unanimously Acceptable Lotteries via Queries
287. ICML 2026Linear Regression with Unknown Truncation Beyond Gaussian Features
288. ICML 2026Linear-Core Surrogates: Smooth Loss Functions with Linear Rates for Classification and Structured Prediction
289. ICML 2026Low-dimensional topology of deep neural networks
290. ICML 2026Matroid Algorithms Under Size-Sensitive Independence Oracles
291. ICML 2026Mean-Shift PCA by Knockoff Mean
292. ICML 2026Mixtures of geodesic factor analyzers on Riemannian homogeneous spaces
293. ICML 2026Multi-Objective Learning for Diffusion Models: A Statistical Theory under Semi-Supervised Learning
294. ICML 2026Multicalibration Yields Better Matchings
295. ICML 2026Next-Token Prediction and Regret Minimization
296. ICML 2026Non-Adversarial Imitation Learning Provably Free of Compounding Errors: The Value Flow Mechanism
297. ICML 2026Off-Policy Evaluation for Missingness-Aware Policies in MDPs with Rewards Missing Not at Random
298. ICML 2026On Contraction of Sequential and Offset Rademacher Complexities
299. ICML 2026On Expressive Power of Floating-Point Transformers
300. ICML 2026On Learnability and Disambiguation of Multiclass Partial Concept Classes

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.