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

1,350 papers, page 16 of 27

751. ICML 2025Towards a Formal Theory of Representational Compositionality
752. ICML 2025Towards Understanding Catastrophic Forgetting in Two-layer Convolutional Neural Networks
753. ICML 2025Training Dynamics of In-Context Learning in Linear Attention
754. ICML 2025Transfer Q-Learning with Composite MDP Structures
755. ICML 2025Tree-Sliced Wasserstein Distance: A Geometric Perspective
756. ICML 2025Understanding Generalization in Quantum Machine Learning with Margins
757. ICML 2025Understanding the Forgetting of (Replay-based) Continual Learning via Feature Learning: Angle Matters
758. ICML 2025Understanding the Statistical Accuracy-Communication Trade-off in Personalized Federated Learning with Minimax Guarantees
759. ICML 2025Unified Analysis of Continuous Weak Features Learning with Applications to Learning from Missing Data
760. ICML 2025Uniform Mean Estimation for Heavy-Tailed Distributions via Median-of-Means
761. ICML 2025Universal Length Generalization with Turing Programs
762. ICML 2025Weak-to-Strong Generalization Even in Random Feature Networks, Provably
763. ICML 2025Weight matrices compression based on PDB model in deep neural networks
764. ICML 2025When Diffusion Models Memorize: Inductive Biases in Probability Flow of Minimum-Norm Shallow Neural Nets
765. ICML 2025When do neural networks learn world models?
766. NeurIPS 2025A Cramér–von Mises Approach to Incentivizing Truthful Data Sharing
767. NeurIPS 2025A Geometric Analysis of PCA
768. NeurIPS 2025A Statistical Theory of Contrastive Learning via Approximate Sufficient Statistics
769. NeurIPS 2025A Theory for Worst-Case vs. Average-Case Guarantees for LLMs
770. NeurIPS 2025A Unified Approach to Submodular Maximization Under Noise
771. NeurIPS 2025A unified framework for establishing the universal approximation of transformer-type architectures
772. NeurIPS 2025Adam Reduces a Unique Form of Sharpness: Theoretical Insights Near the Minimizer Manifold
773. NeurIPS 2025Agnostic Active Learning Is Always Better Than Passive Learning
774. NeurIPS 2025Alternating Gradient Flows: A Theory of Feature Learning in Two-layer Neural Networks
775. NeurIPS 2025An Analytical Theory of Spectral Bias in the Learning Dynamics of Diffusion Models
776. NeurIPS 2025Approximation theory for 1-Lipschitz ResNets
777. NeurIPS 2025Benign Overfitting in Single-Head Attention
778. NeurIPS 2025Better NTK Conditioning: A Free Lunch from (ReLU) Nonlinear Activation in Wide Neural Networks
779. NeurIPS 2025Beyond Benign Overfitting in Nadaraya-Watson Interpolators
780. NeurIPS 2025Block Coordinate Descent for Neural Networks Provably Finds Global Minima
781. NeurIPS 2025Bridging Scales: Spectral Theory Reveals How Local Connectivity Rules Sculpt Global Neural Dynamics in Spatially Extended Networks
782. NeurIPS 2025Can Diffusion Models Disentangle? A Theoretical Perspective
783. NeurIPS 2025Certifying Stability of Reinforcement Learning Policies using Generalized Lyapunov Functions
784. NeurIPS 2025Computable universal online learning
785. NeurIPS 2025Computational Efficiency under Covariate Shift in Kernel Ridge Regression
786. NeurIPS 2025Conformal Inference under High-Dimensional Covariate Shifts via Likelihood-Ratio Regularization
787. NeurIPS 2025Constant Bit-size Transformers Are Turing Complete
788. NeurIPS 2025Constrained Linear Thompson Sampling
789. NeurIPS 2025Contextual Dynamic Pricing with Heterogeneous Buyers
790. NeurIPS 2025Curl Descent : Non-Gradient Learning Dynamics with Sign-Diverse Plasticity
791. NeurIPS 2025Cyclic Counterfactuals under Shift–Scale Interventions
792. NeurIPS 2025Disentangling Hyperedges through the Lens of Category Theory
793. NeurIPS 2025Distributional Training Data Attribution: What do Influence Functions Sample?
794. NeurIPS 2025Do Neural Networks Need Gradient Descent to Generalize? A Theoretical Study
795. NeurIPS 2025Eluder dimension: localise it!
796. NeurIPS 2025Equivariance by Contrast: Identifiable Equivariant Embeddings from Unlabeled Finite Group Actions
797. NeurIPS 2025Exact Expressive Power of Transformers with Padding
798. NeurIPS 2025Exploiting Vocabulary Frequency Imbalance in Language Model Pre-training
799. NeurIPS 2025Fast attention mechanisms: a tale of parallelism
800. NeurIPS 2025Gaussian Process Upper Confidence Bound Achieves Nearly-Optimal Regret in Noise-Free Gaussian Process Bandits

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.