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

393 papers, page 5 of 8

201. NeurIPS 2024Exploring the Precise Dynamics of Single-Layer GAN Models: Leveraging Multi-Feature Discriminators for High-Dimensional Subspace Learning
202. NeurIPS 2024Federated Learning from Vision-Language Foundation Models: Theoretical Analysis and Method
203. NeurIPS 2024FedGTST: Boosting Global Transferability of Federated Models via Statistics Tuning
204. NeurIPS 2024From Dictionary to Tensor: A Scalable Multi-View Subspace Clustering Framework with Triple Information Enhancement
205. NeurIPS 2024Generalizablity of Memorization Neural Network
206. NeurIPS 2024Generalization Analysis for Label-Specific Representation Learning
207. NeurIPS 2024Generalization Bound and Learning Methods for Data-Driven Projections in Linear Programming
208. NeurIPS 2024Generalization Bounds via Conditional $f$-Information
209. NeurIPS 2024Generalization Error Bounds for Two-stage Recommender Systems with Tree Structure
210. NeurIPS 2024Generalization of Hamiltonian algorithms
211. NeurIPS 2024Generalized Eigenvalue Problems with Generative Priors
212. NeurIPS 2024Generalized Tensor Decomposition for Understanding Multi-Output Regression under Combinatorial Shifts
213. NeurIPS 2024Generative Forests
214. NeurIPS 2024Get rich quick: exact solutions reveal how unbalanced initializations promote rapid feature learning
215. NeurIPS 2024Gradual Domain Adaptation via Manifold-Constrained Distributionally Robust Optimization
216. NeurIPS 2024Group-wise oracle-efficient algorithms for online multi-group learning
217. NeurIPS 2024Hardness of Learning Neural Networks under the Manifold Hypothesis
218. NeurIPS 2024Harnessing Multiple Correlated Networks for Exact Community Recovery
219. NeurIPS 2024High-dimensional (Group) Adversarial Training in Linear Regression
220. NeurIPS 2024How Does Black-Box Impact the Learning Guarantee of Stochastic Compositional Optimization?
221. NeurIPS 2024How Far Can Transformers Reason? The Globality Barrier and Inductive Scratchpad
222. NeurIPS 2024How JEPA Avoids Noisy Features: The Implicit Bias of Deep Linear Self Distillation Networks
223. NeurIPS 2024How many classifiers do we need?
224. NeurIPS 2024How Sparse Can We Prune A Deep Network: A Fundamental Limit Perspective
225. NeurIPS 2024How to Boost Any Loss Function
226. NeurIPS 2024How Transformers Utilize Multi-Head Attention in In-Context Learning? A Case Study on Sparse Linear Regression
227. NeurIPS 2024Hyperbolic Embeddings of Supervised Models
228. NeurIPS 2024Identification of Analytic Nonlinear Dynamical Systems with Non-asymptotic Guarantees
229. NeurIPS 2024Implicit Regularization Paths of Weighted Neural Representations
230. NeurIPS 2024Imprecise Label Learning: A Unified Framework for Learning with Various Imprecise Label Configurations
231. NeurIPS 2024Improved Sample Complexity Bounds for Diffusion Model Training
232. NeurIPS 2024Improved Sample Complexity for Multiclass PAC Learning
233. NeurIPS 2024Improving Adaptivity via Over-Parameterization in Sequence Models
234. NeurIPS 2024Improving self-training under distribution shifts via anchored confidence with theoretical guarantees
235. NeurIPS 2024In-Context Learning of a Linear Transformer Block: Benefits of the MLP Component and One-Step GD Initialization
236. NeurIPS 2024In-Context Learning with Transformers: Softmax Attention Adapts to Function Lipschitzness
237. NeurIPS 2024Inductive biases of multi-task learning and finetuning: multiple regimes of feature reuse
238. NeurIPS 2024Information-theoretic Generalization Analysis for Expected Calibration Error
239. NeurIPS 2024Information-theoretic Limits of Online Classification with Noisy Labels
240. NeurIPS 2024Invariant subspaces and PCA in nearly matrix multiplication time
241. NeurIPS 2024Inverse M-Kernels for Linear Universal Approximators of Non-Negative Functions
242. NeurIPS 2024Is Behavior Cloning All You Need? Understanding Horizon in Imitation Learning
243. NeurIPS 2024Is Cross-validation the Gold Standard to Estimate Out-of-sample Model Performance?
244. NeurIPS 2024Is O(log N) practical? Near-Equivalence Between Delay Robustness and Bounded Regret in Bandits and RL
245. NeurIPS 2024Label Noise: Ignorance Is Bliss
246. NeurIPS 2024Language Generation in the Limit
247. NeurIPS 2024Learning a Single Neuron Robustly to Distributional Shifts and Adversarial Label Noise
248. NeurIPS 2024Learning and Transferring Sparse Contextual Bigrams with Linear Transformers
249. NeurIPS 2024Learning from higher-order correlations, efficiently: hypothesis tests, random features, and neural networks
250. NeurIPS 2024Learning from Noisy Labels via Conditional Distributionally Robust Optimization

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.