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

477 papers, page 7 of 10

301. ICLR 2025Progress or Regress? Self-Improvement Reversal in Post-training
302. ICLR 2025Provable Uncertainty Decomposition via Higher-Order Calibration
303. ICLR 2025Provable weak-to-strong generalization via benign overfitting
304. ICLR 2025Quality over Quantity in Attention Layers: When Adding More Heads Hurts
305. ICLR 2025Quantitative Approximation for Neural Operators in Nonlinear Parabolic Equations
306. ICLR 2025Regret Bounds for Episodic Risk-Sensitive Linear Quadratic Regulator
307. ICLR 2025Rethinking Self-Distillation: Label Averaging and Enhanced Soft Label Refinement with Partial Labels
308. ICLR 2025RNNs are not Transformers (Yet): The Key Bottleneck on In-Context Retrieval
309. ICLR 2025Robust Feature Learning for Multi-Index Models in High Dimensions
310. ICLR 2025Robustness Auditing for Linear Regression: To Singularity and Beyond
311. ICLR 2025Satisficing Regret Minimization in Bandits
312. ICLR 2025Second Order Bounds for Contextual Bandits with Function Approximation
313. ICLR 2025Selective Task Group Updates for Multi-Task Optimization
314. ICLR 2025Self-Improvement in Language Models: The Sharpening Mechanism
315. ICLR 2025Semialgebraic Neural Networks: From roots to representations
316. ICLR 2025Separation Power of Equivariant Neural Networks
317. ICLR 2025Sharper Guarantees for Learning Neural Network Classifiers with Gradient Methods
318. ICLR 2025Sketching for Convex and Nonconvex Regularized Least Squares with Sharp Guarantees
319. ICLR 2025State Space Models are Provably Comparable to Transformers in Dynamic Token Selection
320. ICLR 2025Statistical Advantages of Perturbing Cosine Router in Mixture of Experts
321. ICLR 2025Statistical Tractability of Off-policy Evaluation of History-dependent Policies in POMDPs
322. ICLR 2025Stochastic Bandits Robust to Adversarial Attacks
323. ICLR 2025Strong Model Collapse
324. ICLR 2025Task Descriptors Help Transformers Learn Linear Models In-Context
325. ICLR 2025The Breakdown of Gaussian Universality in Classification of High-dimensional Linear Factor Mixtures
326. ICLR 2025The Computational Complexity of Positive Non-Clashing Teaching in Graphs
327. ICLR 2025The Foundations of Tokenization: Statistical and Computational Concerns
328. ICLR 2025Theory on Mixture-of-Experts in Continual Learning
329. ICLR 2025Theory on Score-Mismatched Diffusion Models and Zero-Shot Conditional Samplers
330. ICLR 2025Towards Auto-Regressive Next-Token Prediction: In-context Learning Emerges from Generalization
331. ICLR 2025Towards Generalization Bounds of GCNs for Adversarially Robust Node Classification
332. ICLR 2025Towards Understanding the Universality of Transformers for Next-Token Prediction
333. ICLR 2025Tractable Multi-Agent Reinforcement Learning through Behavioral Economics
334. ICLR 2025Trained Transformer Classifiers Generalize and Exhibit Benign Overfitting In-Context
335. ICLR 2025Training Nonlinear Transformers for Chain-of-Thought Inference: A Theoretical Generalization Analysis
336. ICLR 2025Transformer Learns Optimal Variable Selection in Group-Sparse Classification
337. ICLR 2025Transformers are Universal In-context Learners
338. ICLR 2025Transformers Handle Endogeneity in In-Context Linear Regression
339. ICLR 2025Transformers Learn to Implement Multi-step Gradient Descent with Chain of Thought
340. ICLR 2025Transformers Provably Learn Two-Mixture of Linear Classification via Gradient Flow
341. ICLR 2025Transformers Provably Solve Parity Efficiently with Chain of Thought
342. ICLR 2025U-Nets as Belief Propagation: Efficient Classification, Denoising, and Diffusion in Generative Hierarchical Models
343. ICLR 2025Understanding Factual Recall in Transformers via Associative Memories
344. ICLR 2025uniINF: Best-of-Both-Worlds Algorithm for Parameter-Free Heavy-Tailed MABs
345. ICLR 2025When does compositional structure yield compositional generalization? A kernel theory.
346. ICLR 2025When is Task Vector Provably Effective for Model Editing? A Generalization Analysis of Nonlinear Transformers
347. ICLR 2025When narrower is better: the narrow width limit of Bayesian parallel branching neural networks
348. ICLR 2024A Black-box Approach for Non-stationary Multi-agent Reinforcement Learning
349. ICLR 2024A path-norm toolkit for modern networks: consequences, promises and challenges
350. ICLR 2024A Policy Gradient Method for Confounded POMDPs

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.