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
[PDF]How Multi-Head Transformers Specialize for In-Context Learning: A Theoretical Analysis of Training Dynamics and Generalization▲2843.00
[PDF]Quantized Keys Steal Attention: Bias Correction for KV-Cache Compression in Video Diffusion▲5103.00
[PDF]Softmax Attention as a Covariance Readout: A Unified View of In-Context Learning and Repetition▲8.9k2.75
[PDF]Stagewise Learning of Balanced Orthogonal Mixtures by Finite-Width Two-Head Transformers▼1351.33
[PDF]Transformers Can Implement Preconditioned Richardson Iteration for In-Context Gaussian Kernel Regression▲1.7k3.33