Related papers: FastAttention: Extend FlashAttention2 to NPUs and …

Scaling Transformers to ultra-long contexts is bottlenecked by the $O(n^2 d)$ cost of self-attention. Existing methods reduce this cost along the sequence axis through local windows, kernel approximations, or token-level sparsity, but these…

Large Language Models (LLMs) demonstrate substantial potential across a diverse array of domains via request serving. However, as trends continue to push for expanding context sizes, the autoregressive nature of LLMs results in highly…

TiledAttention is a scaled dot-product attention (SDPA) forward operator for SDPA research on NVIDIA GPUs. Implemented in cuTile Python (TileIR) and exposed as a PyTorch-callable function, it is easier to modify than low-level CUDA…

Inference with Transformer-based Large Language Models (LLMs) on long sequences is both costly and slow due to the quadratic complexity of the self-attention mechanism. We introduce Star Attention, a two-phase block-sparse approximation…

The rapid development of large language models (LLM) has greatly enhanced everyday applications. While many FPGA-based accelerators, with flexibility for fine-grained data control, exhibit superior speed and energy efficiency compared to…

Transformers are mostly relying on softmax attention, which introduces quadratic complexity with respect to sequence length and remains a major bottleneck for efficient inference. Prior work on linear or hybrid attention typically replaces…

In this work, we present FastAV, the first token pruning framework tailored for audio-visual large language models (AV-LLMs). While token pruning has been actively explored in standard large language models (LLMs) and vision-language models…

The efficiency of attention is critical because its time complexity grows quadratically with sequence length. SageAttention2 addresses this by utilizing quantization to accelerate matrix multiplications (Matmul) in attention. To further…

Attention-based neural networks have become pervasive in many AI tasks. Despite their excellent algorithmic performance, the use of the attention mechanism and feed-forward network (FFN) demands excessive computational and memory resources,…

Reducing the key-value (KV) cache burden in Large Language Models (LLMs) significantly accelerates inference. Dynamically selecting critical KV caches during decoding helps maintain performance. Existing methods use random linear hashing to…

Multi-scale deformable attention (MSDA) is a flexible and powerful feature extraction mechanism for visual tasks, but its random-access grid sampling strategy poses significant optimization challenges, especially on domain-specific…

Large vision-language models (VLMs) enable joint processing of text and images. However, incorporating vision data significantly increases the prompt length, resulting in a longer time to first token (TTFT). This bottleneck can be…

Operator fusion has become a key optimization for deep learning, which combines multiple deep learning operators to improve data reuse and reduce global memory transfers. However, existing tensor compilers struggle to fuse complex reduction…

The growing demand for long-context inference capabilities in Large Language Models (LLMs) has intensified the computational and memory bottlenecks inherent to the self-attention mechanism. To address this challenge, we introduce BLASST, a…

The attention layer, a core component of Transformer-based LLMs, brings out inefficiencies in current GPU systems due to its low operational intensity and the substantial memory requirements of KV caches. We propose a High-bandwidth…

Modern diffusion models, particularly those utilizing a Transformer-based UNet for denoising, rely heavily on self-attention operations to manage complex spatial relationships, thus achieving impressive generation performance. However, this…

Large Language Models (LLMs) encounter severe memory inefficiencies during long-context inference due to conventional handling of key-value (KV) caches. In this work, we introduce a novel integration of PagedAttention with PyTorch's…

Attention-based deep networks have been successfully applied on textual data in the field of NLP. However, their application on protein sequences poses additional challenges due to the weak semantics of the protein words, unlike the plain…

Large-scale diffusion models such as FLUX (12B parameters) and Stable Diffusion 3 (8B parameters) require multi-GPU parallelism for efficient inference. Unified Sequence Parallelism (USP), which combines Ulysses and Ring attention…

Existing methods for training LLMs on long-sequence data, such as Tensor Parallelism and Context Parallelism, exhibit low Model FLOPs Utilization as sequence lengths and number of GPUs increase, especially when sequence lengths exceed 1M…