Raja Giryes

Entropy-preserving reinforcement learning

Adaptive Guidance Scaling for Posterior Diffusion-based Sampling

On the Recall Scaling Laws in Mamba: A Theoretical and Mechanistic Study via Hashing

Revisiting Glorot Initialization for Long-Range Linear Recurrences

Overflow Prevention Enhances Long-Context Recurrent LLMs

Smooth Regularization for Efficient Video Recognition

LiveXiv - A Multi-Modal live benchmark based on Arxiv papers content

DeciMamba: Exploring the Length Extrapolation Potential of Mamba

On the Relation Between Linear Diffusion and Power Iteration

Comparison Visual Instruction Tuning