Related papers: IntAttention: A Fully Integer Attention Pipeline f…

We prove that with linear transformations, both (i) two-layer self-attention and (ii) one-layer self-attention followed by a softmax function are universal approximators for continuous sequence-to-sequence functions on compact domains. Our…

We introduce Performers, Transformer architectures which can estimate regular (softmax) full-rank-attention Transformers with provable accuracy, but using only linear (as opposed to quadratic) space and time complexity, without relying on…

Long-Context Transformer Models (LCTMs) are vital for real-world applications but suffer high computational costs due to attention's quadratic complexity. Block-sparse attention mitigates this by focusing computation on critical regions,…

Transformers' quadratic complexity with respect to the input sequence length has motivated a body of work on efficient sparse approximations to softmax. An alternative path, used by entmax transformers, consists of having built-in exact…

Vision transformers (ViTs) have pushed the state-of-the-art for visual perception tasks. The self-attention mechanism underpinning the strength of ViTs has a quadratic complexity in both computation and memory usage. This motivates the…

We present FIT: a transformer-based architecture with efficient self-attention and adaptive computation. Unlike original transformers, which operate on a single sequence of data tokens, we divide the data tokens into groups, with each group…

Attention is a core component of transformer architecture, whether encoder-only, decoder-only, or encoder-decoder model. However, the standard softmax attention often produces noisy probability distribution, which can impair effective…

Diffusion transformers have demonstrated remarkable capabilities in generating videos. However, their practical deployment is severely constrained by high memory usage and computational cost. Post-Training Quantization provides a practical…

To help address the growing demand for ever-longer sequence lengths in transformer models, Liu et al. recently proposed Ring Attention, an exact attention algorithm capable of overcoming per-device memory bottle- necks by distributing…

Transformers and large language models (LLMs) have revolutionized machine learning, with attention mechanisms at the core of their success. As the landscape of attention variants expands, so too do the challenges of optimizing their…

Recently, transformer-based methods have dominated 3D instance segmentation, where mask attention is commonly involved. Specifically, object queries are guided by the initial instance masks in the first cross-attention, and then iteratively…

We propose a communication-efficient collaborative inference framework in the domain of edge inference, focusing on the efficient use of vision transformer (ViT) models. The partitioning strategy of conventional collaborative inference…

The softmax content-based attention mechanism has proven to be very beneficial in many applications of recurrent neural networks. Nevertheless it suffers from two major computational limitations. First, its computations for an attention…

Attention efficiency is critical to large language model (LLM) inference. While prior advances optimize attention execution for individual requests (e.g., FlashAttention), production LLM serving relies on batching requests with highly…

Attention architectures are widely used; they recently gained renewed popularity with Transformers yielding a streak of state of the art results. Yet, the geometrical implications of softmax-attention remain largely unexplored. In this work…

We propose a simple modification to the conventional attention mechanism applied by Transformers: Instead of quantifying pairwise query-key similarity with scaled dot-products, we quantify it with the logarithms of scaled dot-products of…

Widely adopted in modern Vision Transformer designs, Softmax attention can effectively capture long-range visual information; however, it incurs excessive computational cost when dealing with high-resolution inputs. In contrast, linear…

Gating mechanisms have been widely utilized, from early models like LSTMs and Highway Networks to recent state space models, linear attention, and also softmax attention. Yet, existing literature rarely examines the specific effects of…

We introduce softpick, a rectified, not sum-to-one, drop-in replacement for softmax in transformer attention mechanisms that eliminates attention sink and massive activations. Our experiments with 340M and 1.8B parameter models demonstrate…

Catastrophic forgetting poses a fundamental challenge in continual learning, particularly when models are quantized for deployment efficiency. We systematically investigate the interplay between quantization precision (FP16, INT8, INT4) and…