{"total":21,"rows":[{"id":"2406.06467","title":"How Far Can Transformers Reason? The Globality Barrier and Inductive Scratchpad","created":"2024-06-10","cited_by_count":36,"area":null},{"id":"2506.18110","title":"RL for Reasoning by Adaptively Revealing Rationales","created":"2025-06-22","cited_by_count":16,"area":null},{"id":"2409.18061","title":"Optimal Protocols for Continual Learning via Statistical Physics and Control Theory","created":"2024-09-26","cited_by_count":14,"area":null},{"id":"2603.04964","title":"Replaying pre-training data improves fine-tuning","created":"2026-03-05","cited_by_count":13,"area":null},{"id":"2406.01589","title":"Tilting the Odds at the Lottery: the Interplay of Overparameterisation and Curricula in Neural Networks","created":"2024-06-03","cited_by_count":9,"area":null},{"id":"2406.11110","title":"How Neural Networks Learn the Support is an Implicit Regularization Effect of SGD","created":"2024-06-17","cited_by_count":8,"area":null},{"id":"2502.06443","title":"Low-dimensional Functions are Efficiently Learnable under Randomly Biased Distributions","created":"2025-02-10","cited_by_count":8,"area":null},{"id":"2510.25108","title":"Shift is Good: Mismatched Data Mixing Improves Test Performance","created":"2025-10-29","cited_by_count":5,"area":null},{"id":"2410.03569","title":"Making Hard Problems Easier with Custom Data Distributions and Loss Regularization: A Case Study in Modular Arithmetic","created":"2024-10-04","cited_by_count":4,"area":null},{"id":"2502.07553","title":"Transformers Provably Learn Sparse XOR with Polylogarithmic Parameters","created":"2025-02-11","cited_by_count":3,"area":null},{"id":"2507.07907","title":"A statistical physics framework for optimal learning","created":"2025-07-10","cited_by_count":3,"area":null},{"id":"2602.08907","title":"Positive Distribution Shift as a Framework for Understanding Tractable Learning","created":"2026-02-09","cited_by_count":3,"area":null},{"id":"2312.05379","title":"Exploring Parity Challenges in Reinforcement Learning through Curriculum Learning with Noisy Labels","created":"2023-12-08","cited_by_count":1,"area":null},{"id":"2410.24050","title":"A Mechanistic Study of Transformers Training Dynamics","created":"2024-10-31","cited_by_count":1,"area":null},{"id":"2605.10237","title":"The Benefits of Temporal Correlations: SGD Learns k-Juntas from Random Walks Efficiently","created":"2026-05-11","cited_by_count":1,"area":null},{"id":"2606.20937","title":"Learning through Internalization","created":"2026-06-18","cited_by_count":1,"area":null},{"id":"2501.02362","title":"Easing Optimization Paths: a Circuit Perspective","created":"2025-01-04","cited_by_count":0,"area":null},{"id":"2604.22951","title":"The Power of Power Law: Asymmetry Enables Compositional Reasoning","created":"2026-04-24","cited_by_count":0,"area":null},{"id":"2605.07648","title":"Learning Large-Scale Modular Addition with an Auxiliary Modulus","created":"2026-05-08","cited_by_count":0,"area":null},{"id":"2605.20314","title":"Less Data, Faster Training: repeating smaller datasets speeds up learning via sampling biases","created":"2026-05-19","cited_by_count":0,"area":null},{"id":"2607.18305","title":"The Information Shadow: Measuring Structural Limits on What Language Models Can Learn","created":"2026-07-17","cited_by_count":0,"area":null}],"next_cursor":null}