相关论文: Flex Attention: A Programming Model for Generating…

Long-context modeling is a pivotal capability for Large Language Models, yet the quadratic complexity of attention remains a critical bottleneck, particularly during the compute-intensive prefilling phase. While various sparse attention…

Transformers based on the attention mechanism have achieved impressive success in various areas. However, the attention mechanism has a quadratic complexity, significantly impeding Transformers from dealing with numerous tokens and scaling…

Deploying Transformer models on edge devices is limited by latency and energy budgets. While INT8 quantization effectively accelerates the primary matrix multiplications, it exposes the softmax as the dominant bottleneck. This stage incurs…

Transformer-based models, such as BERT and GPT, have been widely adopted in natural language processing (NLP) due to their exceptional performance. However, recent studies show their vulnerability to textual adversarial attacks where the…

Transformers have emerged as the architecture of choice for many state-of-the-art AI models, showcasing exceptional performance across a wide range of AI applications. However, the memory demands imposed by Transformers limit their ability…

There has recently been significant interest in training reinforcement learning (RL) agents in vision-based environments. This poses many challenges, such as high dimensionality and the potential for observational overfitting through…

Pre-trained language models (PLM) have demonstrated their effectiveness for a broad range of information retrieval and natural language processing tasks. As the core part of PLM, multi-head self-attention is appealing for its ability to…

We propose Axial Transformers, a self-attention-based autoregressive model for images and other data organized as high dimensional tensors. Existing autoregressive models either suffer from excessively large computational resource…

Transformer-based large language models (LLMs) have demonstrated remarkable performance across a wide range of real-world tasks, but their inference cost remains prohibitively high due to the quadratic complexity of attention and the memory…

Accommodating long sequences efficiently in autoregressive Transformers, especially within an extended context window, poses significant challenges due to the quadratic computational complexity and substantial KV memory requirements…

Autoregressive Transformers are strong language models but incur O(T) complexity during per-token generation due to the self-attention mechanism. Recent work proposes kernel-based methods to approximate causal self-attention by replacing it…

Transformer is a powerful model for text understanding. However, it is inefficient due to its quadratic complexity to input sequence length. Although there are many methods on Transformer acceleration, they are still either inefficient on…

The neural attention mechanism has been incorporated into deep neural networks to achieve state-of-the-art performance in various domains. Most such models use multi-head self-attention which is appealing for the ability to attend to…

In humans, Attention is a core property of all perceptual and cognitive operations. Given our limited ability to process competing sources, attention mechanisms select, modulate, and focus on the information most relevant to behavior. For…

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…

FlashAttention-style online softmax enables exact attention computation with linear memory by streaming score tiles through on-chip memory and maintaining a running maximum and normalizer. However, as attention kernels approach peak…

Vision transformers using self-attention or its proposed alternatives have demonstrated promising results in many image related tasks. However, the underpinning inductive bias of attention is not well understood. To address this issue, this…

Large Reasoning Models (LRMs) have shown promising accuracy improvements on complex problem-solving tasks. While these models have attained high accuracy by leveraging additional computation at test time, they need to generate long…

We introduced a {\it working memory} augmented adaptive controller in our recent work. The controller uses attention to read from and write to the working memory. Attention allows the controller to read specific information that is relevant…

Following the success of dot-product attention in Transformers, numerous approximations have been recently proposed to address its quadratic complexity with respect to the input length. While these variants are memory and compute efficient,…