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 9 of 10

401. NeurIPS 2026Estimating Implicit Regularization in Deep Learning
402. NeurIPS 2026Estimating the expected output of wide random MLPs more efficiently than sampling
403. NeurIPS 2026Estimation of the Label-Noise Transition Matrix with Performance Guarantees via Selective Classification
404. NeurIPS 2026Fast Learning Rates for Physics-Informed Kernel Methods
405. NeurIPS 2026Finite-Sample Performance of Gradient Descent in Logistic Regression with Gaussian Design
406. NeurIPS 2026First-Order Regret for Online Convex Optimization with Memory and Online Nonstochastic Control
407. NeurIPS 2026Fixed Universal Transformers
408. NeurIPS 2026Foundations of Categorical Equivariant Deep Learning
409. NeurIPS 2026From POMDP Theory to Deep RL with Particle Filters
410. NeurIPS 2026Generalization Analysis of Biased Stochastic Gradient Methods for Minimax Problems
411. NeurIPS 2026Generalization Bounds for Neural Networks with Sparse Connectivity
412. NeurIPS 2026Generating in the Limit with Infinitely Many Hallucinations
413. NeurIPS 2026Global Convergence in Deep Networks via the Second Law of Thermodynamics
414. NeurIPS 2026Grokking or Glitching? How Low-Precision Drives Slingshot Loss Spikes
415. NeurIPS 2026High dimensional theory of two-phase optimizers
416. NeurIPS 2026High-arity Sample Compression
417. NeurIPS 2026How to Scale Mixture-of-Experts: From muP to the Maximally Scale-Stable Parameterization
418. NeurIPS 2026Imperfect Influence, Reliable Rankings: A Theory of TRAK for Data Attribution
419. NeurIPS 2026Improved Algorithms for Online Classification with Surrogate Losses
420. NeurIPS 2026Inference and Uncertainty Quantification for Streaming $r$-PCA
421. NeurIPS 2026Inference Time Nash Alignment
422. NeurIPS 2026Inferential Theory of Learning as a Framework for Test-Time Computation in Foundation Models
423. NeurIPS 2026InfoFlow: A Framework for Multi-Layer Transformer Analysis
424. NeurIPS 2026Information Loss and Disparate Effects in Network Embeddings
425. NeurIPS 2026Learning Acceptable Lotteries via Queries: Minimizing Aggregated Violation Distances
426. NeurIPS 2026Learning Chance-Constrained MDPs with Bellman Distributional Certificates
427. NeurIPS 2026Learning Lotteries with Minimal Violations from Membership Queries
428. NeurIPS 2026Learning to Audit ML Models with Theory of Mind
429. NeurIPS 2026Less Structure is More: Minimal Representations for Supervised Learning
430. NeurIPS 2026Linear Contextual Bandits with Quasi-Optimism
431. NeurIPS 2026Linear Ensemble Sampling with Fewer Ensembles
432. NeurIPS 2026Looped Transformers with Layer Normalization Provably Learn the Power Method
433. NeurIPS 2026Memory Determines Learning Direction: A Theory of Gradient-Based Optimization in State Space Models
434. NeurIPS 2026Minimax Optimal Early-Stopped Gradient Descent for Gaussian Mixture Classification
435. NeurIPS 2026MinMax Recurrent Neural Cascades
436. NeurIPS 2026Model Collapse is a Singular Complexity Trajectory
437. NeurIPS 2026Multi-Armed Bandits With Best-Action Queries
438. NeurIPS 2026Multiple Descent of Generalization Curve for Optimally Regularized Ridge Regression
439. NeurIPS 2026Near-Optimal Best-of-Both-Worlds Algorithms for Decoupled Exploration and Exploitation in Multi-armed Bandits
440. NeurIPS 2026Network Intervention by Polling Strategic Agents
441. NeurIPS 2026Nonparametric In-Context Learning under Growing Geometric Complexity: Minimax Optimality and Local Geometry-Adaptivity of Transformers
442. NeurIPS 2026On Concentration Inequalities for Sampling without Replacement
443. NeurIPS 2026On the Complexity of Preference-Based Bandits
444. NeurIPS 2026One-Layer Transformers Provably Learn In-Context K-Nearest Neighbor Prediction with Chain-of-Thought
445. NeurIPS 2026Online Directional Regression for Streaming Sufficient Dimension Reduction
446. NeurIPS 2026Online Resource Allocation With General Constraints
447. NeurIPS 2026Online Set Learning from Precision and Recall Feedback
448. NeurIPS 2026Optimal Dimension-Free Sampling for Regularized Classification
449. NeurIPS 2026PAC-Bayesian Bounds for Learning Partially Observed Stochastic Linear Time-Invariant State-Space Systems with Inputs and Sub-Gaussian Noise
450. NeurIPS 2026Parameter symmetries determine representational geometry in overparameterized nonlinear networks

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.