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

389 papers, page 6 of 8

251. ICML 2025Representative Language Generation
252. ICML 2025Rethinking Benign Overfitting in Two-Layer Neural Networks
253. ICML 2025Retraining with Predicted Hard Labels Provably Increases Model Accuracy
254. ICML 2025Revisiting the Predictability of Performative, Social Events
255. ICML 2025Sample-Optimal Agnostic Boosting with Unlabeled Data
256. ICML 2025Sharp Optimality of Simple, Plug-in Estimation of the Fisher Information of a Smoothed Density
257. ICML 2025Sort Before You Prune: Improved Worst-Case Guarantees of the DiskANN Family of Graphs
258. ICML 2025Stability and Generalization Analysis of Decentralized SGD: Sharper Bounds Beyond Lipschitzness and Smoothness
259. ICML 2025Statistical Query Hardness of Multiclass Linear Classification with Random Classification Noise
260. ICML 2025System-Aware Unlearning Algorithms: Use Lesser, Forget Faster
261. ICML 2025The Butterfly Effect: Neural Network Training Trajectories Are Highly Sensitive to Initial Conditions
262. ICML 2025The Empirical Mean is Minimax Optimal for Local Glivenko-Cantelli
263. ICML 2025The impact of uncertainty on regularized learning in games
264. ICML 2025The Power of Random Features and the Limits of Distribution-Free Gradient Descent
265. ICML 2025The Role of Randomness in Stability
266. ICML 2025Theoretical Limitations of Ensembles in the Age of Overparameterization
267. ICML 2025Tight and Fast Bounds for Multi-Label Learning
268. ICML 2025Toward a Unified Theory of Gradient Descent under Generalized Smoothness
269. ICML 2025Towards a Formal Theory of Representational Compositionality
270. ICML 2025Towards Understanding Catastrophic Forgetting in Two-layer Convolutional Neural Networks
271. ICML 2025Training Dynamics of In-Context Learning in Linear Attention
272. ICML 2025Transfer Q-Learning with Composite MDP Structures
273. ICML 2025Tree-Sliced Wasserstein Distance: A Geometric Perspective
274. ICML 2025Understanding Generalization in Quantum Machine Learning with Margins
275. ICML 2025Understanding the Forgetting of (Replay-based) Continual Learning via Feature Learning: Angle Matters
276. ICML 2025Understanding the Statistical Accuracy-Communication Trade-off in Personalized Federated Learning with Minimax Guarantees
277. ICML 2025Unified Analysis of Continuous Weak Features Learning with Applications to Learning from Missing Data
278. ICML 2025Uniform Mean Estimation for Heavy-Tailed Distributions via Median-of-Means
279. ICML 2025Universal Length Generalization with Turing Programs
280. ICML 2025Weak-to-Strong Generalization Even in Random Feature Networks, Provably
281. ICML 2025Weight matrices compression based on PDB model in deep neural networks
282. ICML 2025When Diffusion Models Memorize: Inductive Biases in Probability Flow of Minimum-Norm Shallow Neural Nets
283. ICML 2025When do neural networks learn world models?
284. NeurIPS 2025A Cramér–von Mises Approach to Incentivizing Truthful Data Sharing
285. NeurIPS 2025A Geometric Analysis of PCA
286. NeurIPS 2025A Statistical Theory of Contrastive Learning via Approximate Sufficient Statistics
287. NeurIPS 2025A Theory for Worst-Case vs. Average-Case Guarantees for LLMs
288. NeurIPS 2025A Unified Approach to Submodular Maximization Under Noise
289. NeurIPS 2025A unified framework for establishing the universal approximation of transformer-type architectures
290. NeurIPS 2025Adam Reduces a Unique Form of Sharpness: Theoretical Insights Near the Minimizer Manifold
291. NeurIPS 2025Agnostic Active Learning Is Always Better Than Passive Learning
292. NeurIPS 2025Alternating Gradient Flows: A Theory of Feature Learning in Two-layer Neural Networks
293. NeurIPS 2025An Analytical Theory of Spectral Bias in the Learning Dynamics of Diffusion Models
294. NeurIPS 2025Approximation theory for 1-Lipschitz ResNets
295. NeurIPS 2025Benign Overfitting in Single-Head Attention
296. NeurIPS 2025Better NTK Conditioning: A Free Lunch from (ReLU) Nonlinear Activation in Wide Neural Networks
297. NeurIPS 2025Beyond Benign Overfitting in Nadaraya-Watson Interpolators
298. NeurIPS 2025Block Coordinate Descent for Neural Networks Provably Finds Global Minima
299. NeurIPS 2025Bridging Scales: Spectral Theory Reveals How Local Connectivity Rules Sculpt Global Neural Dynamics in Spatially Extended Networks
300. NeurIPS 2025Can Diffusion Models Disentangle? A Theoretical Perspective

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.