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 3 of 7

101. ICML 2026On the Accuracy of Newton Step and Influence Function Data Attributions
102. ICML 2026On the Emergence of Implicit Curriculum in RLVR Learning Dynamics
103. ICML 2026On the Existence of Consistent Adversarial Attacks in High-Dimensional Linear Classification
104. ICML 2026On the Power of Source Screening for Learning Shared Feature Extractors
105. ICML 2026On the Sharp Input-Output Analysis of Nonlinear Systems under Adversarial Attacks
106. ICML 2026On the Theory of Continual Learning with Gradient Descent for Neural Networks
107. ICML 2026On the Variability of Concept Activation Vectors
108. ICML 2026Online Learning with Recency: Algorithms for Sliding-window Streaming Multi-armed Bandits
109. ICML 2026Optimal Learning from Label Proportions with General Loss Functions
110. ICML 2026Optimal Stopping in Latent Diffusion Models
111. ICML 2026Outcome-Aware Spectral Feature Learning for Instrumental Variable Regression
112. ICML 2026PAC-Bayesian Reinforcement Learning Trains Generalizable Policies
113. ICML 2026PCA of Probability Measures: Sparse and Dense Sampling Regimes
114. ICML 2026Performative Learning Theory
115. ICML 2026Positive Distribution Shift as a Framework for Understanding Tractable Learning
116. ICML 2026Provable Bounds for the Learnability of Sample-Compressible Families from Noisy Samples
117. ICML 2026Provable Sample Efficiency of Curriculum Post-Training for Transformer Reasoning
118. ICML 2026Provably Data-driven Multiple Hyper-parameter Tuning with Structured Loss Function
119. ICML 2026Provably Label-Efficient Conformal Prediction
120. ICML 2026Provably Learning Attention with Queries
121. ICML 2026Rational Neural Networks have Expressivity Advantages
122. ICML 2026Rational Transductors
123. ICML 2026Realizable Bayes-Consistency for General Metric Losses
124. ICML 2026Refined Analysis of Entropy-Regularized Actor-Critic
125. ICML 2026Representation Learning for Equivariant Inference with Guarantees
126. ICML 2026Revisiting the Volume Hypothesis
127. ICML 2026Sample Complexity Bounds for Robust Mean Estimation with Mean-Shift Contamination
128. ICML 2026SAQNN: Spectral Adaptive Quantum Neural Network as a Universal Approximator
129. ICML 2026Scalable and Stable Estimation of Amari $\alpha$-Divergence using Random Fourier Features
130. ICML 2026Scaling Laws for Precision in High-Dimensional Linear Regression
131. ICML 2026Semi-knockoffs: a model-agnostic conditional independence testing method with finite-sample guarantees
132. ICML 2026Sharp description of local minima in the loss landscape of high-dimensional two-layer ReLU neural networks
133. ICML 2026Sharp Empirical Bernstein Inequalities for the Variance of Bounded Random Variables
134. ICML 2026Sharp Inequalities between Total Variation and Hellinger Distances for Gaussian Mixtures
135. ICML 2026Sharper Generalization Guarantees for Asynchronous SGD: Beyond Lipschitzness, Smoothness and Data Homogeneity
136. ICML 2026Sparse Regression with $\ell_0$ Constraints for $\alpha$-Mixing Time Series: Algorithms and Guarantees
137. ICML 2026Stability and Generalization of Nonconvex Optimization with Heavy-Tailed Noise
138. ICML 2026Stability beyond Bounded Differences: Sharp Generalization Bounds under Finite $L_p$ Moments
139. ICML 2026Statistical Consistency and Generalization of Contrastive Representation Learning
140. ICML 2026Statistically Undetectable Backdoors in Deep Neural Networks
141. ICML 2026The Catastrophic Failure of *the* k-Means Algorithm in High Dimensions, and How Hartigan's Algorithm Avoids It
142. ICML 2026The Data Manifold under the Microscope
143. ICML 2026The Devil is in the Condition Numbers: Why is GLU Better than non-GLU Structure?
144. ICML 2026The Entropic Signature of Class Speciation in Diffusion Models
145. ICML 2026The Fisher Dimension: Instance-Dependent Complexity for Causal Discovery
146. ICML 2026The Implicit Bias of Steepest Descent with Mini-batch Stochastic Gradient
147. ICML 2026The Interplay Between Interpolation and Aggregation in Regression: Optimal Sample Complexity
148. ICML 2026The Optimal Sample Complexity of Linear Contracts
149. ICML 2026The Stability of Singular Distribution: A Spectral Perspective on the Two-Phase Dynamics of Language Model Pre-training
150. ICML 2026Theoretical Investigation on Inductive Bias of Isolation Forest

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.