NeurIPS 2026 reviews are up: reviews, scores, rebuttals and meta-reviews for 7,877 papers. See the highest rated
acceptodds: the papers of ICLR 2027
1,181 results for “area:"learning theory"” in ICLR 2027. 193 withdrawn or desk rejected listed apart. clearsort:relevanceratingnewestmost read
[PDF]Overfitting of Spectral Gradient Descent: How Matrix Geometry shapes Generalization and Implicit Bias▲5103.00
[PDF]Self-Evolution Should Be Stable, Not Frozen: Lyapunov-Gated Recursive Self-Improvement▼4.3k2.00
[PDF]Attractor Degeneracy: A Theoretical Analysis of Equivalent Memory Classes in Hopfield Neural Networks▼1.2k1.67
[PDF]Sampling the Shell: From Exponential to Constant Sample Complexity in Negative-Sampling Anomaly Detection▼5.8k2.33
[PDF]Implicit Regularization of Bregman Proximal Point Algorithm on Separable Data: Sharp Alignment Bounds and Kullback-Leibler Geometries▼1.2k1.67
[PDF]Instance-Dependent Auto-Exploration for Tabular Online Discounted Reinforcement Learning▲3.0k2.33
[PDF]Theoretical Analysis of Multimodal Contrastive Learning: When and How Hard Negatives Help▼4.3k2.00
[PDF]Route-Conditioned Prediction Extrapolation: Dependent-Data Directions, Persistence, and Certification▼4.3k2.00
[PDF]Model Collapse Is Avoidable Even as Real Data Vanishes: A Sharp Criterion for Iterative Gaussian Estimation▲3952.67
[PDF]Online Geometric Optimization via Projection, Hierarchical Discretization, and Convex Relaxation▼4.5k2.25
[PDF]AI Harness: Certification under Proposal-Conditioned Information for Foundation-Model Agents▼1361.33
Withdrawn or desk rejected (193)
[PDF]Curvature Controlled Mixing on the Interpolation Manifold: Sharp Cutoffs in Late-Phase Training DynamicsDesk rejected
[PDF]Exact Recovery Thresholds for Weighted Data Selection in Vector-Valued Linear RegressionDesk rejected