arXiv AI
Sinkhorn doubly stochastic attention rank decay analysis
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Machine Learning
Sep 10
Convergent Stochastic Training of Multi-Headed Attention and Understanding LoRA
The paper establishes rigorous trainability results for multi-headed attention layers and Low Rank Adaptation (LoRA) models under stochastic training methods. By proving that the empirical regression loss induces a Poincaré inequality with constants independent of data dimension for LoRA and independent of head dimensions for multi-head attention, the authors show that a stochastic differential equation mimicking SGD converges to the loss minima. These results hold without assumptions on data or model size, providing the first theoretical guarantees for training such architectures.
arXiv AI
Jun 29
IWP: Token Pruning as Implicit Weight Pruning in Large Vision Language Models
arXiv:2604. 00757v2 Announce Type: replace-cross Abstract: Large Vision Language Models show impressive performance across image and video understanding tasks, yet their computational cost grows rapidly with the number of visual tokens.
arXiv Machine Learning
Jul 10
Threshold Differential Attention for Sink-Free, Ultra-Sparse, and Non-Dispersive Language Modeling
arXiv:2601. 12145v3 Announce Type: replace Abstract: Softmax attention struggles with long contexts due to structural limitations: the strict sum-to-one constraint forces attention sinks on irrelevant tokens, and probability mass disperses as sequence lengths increase.
arXiv Machine Learning
Jun 4
Customizing the Inductive Biases of Softmax Attention using Structured Matrices
arXiv:2509. 07963v2 Announce Type: replace Abstract: The core component of attention is the scoring function, which transforms the inputs into low-dimensional queries and keys and takes the dot product of each pair.
arXiv AI
Jul 23
ELSAA: Efficient Low-Rank and Sparse Attention Approximation for Training Transformers
arXiv:2607. 20214v1 Announce Type: cross Abstract: The quadratic $N\times N$ attention score matrix remains a central obstacle to extending Transformers to longer input lengths.
arXiv Machine Learning
Sep 30
On State Reduction in Linear Attention
arXiv:2602.04852v3 Announce Type: replace Abstract: Linear attention offers a computationally efficient yet expressive alternative to softmax attention. However, recent empirical results indicate tha...