Related papers: SDBERT: SparseDistilBERT, a faster and smaller BER…

Knowledge distillation (KD) is widely used for compressing a teacher model to a smaller student model, reducing its inference cost and memory footprint while preserving model capabilities. However, current KD methods for auto-regressive…

Transformers are powerful sequence models, but require time and memory that grows quadratically with the sequence length. In this paper we introduce sparse factorizations of the attention matrix which reduce this to $O(n \sqrt{n})$. We also…

The Transformer model is widely successful on many natural language processing tasks. However, the quadratic complexity of self-attention limit its application on long text. In this paper, adopting a fine-to-coarse attention mechanism on…

Diffusion Transformers, particularly for video generation, achieve remarkable quality but suffer from quadratic attention complexity, leading to prohibitive latency. Existing acceleration methods face a fundamental trade-off: dynamically…

How can we efficiently compress a model while maintaining its performance? Knowledge Distillation (KD) is one of the widely known methods for model compression. In essence, KD trains a smaller student model based on a larger teacher model…

Diffusion Transformers (DiT) have become the de-facto model for generating high-quality visual content like videos and images. A huge bottleneck is the attention mechanism where complexity scales quadratically with resolution and video…

Since their introduction the Trasformer architectures emerged as the dominating architectures for both natural language processing and, more recently, computer vision applications. An intrinsic limitation of this family of "fully-attentive"…

Scaling Diffusion Transformer (DiT) inference via sequence parallelism is critical for reducing latency in visual generation, but is severely hampered by workload imbalance when applied to models employing block-wise sparse attention. The…

Recently, BERT has become an essential ingredient of various NLP deep models due to its effectiveness and universal-usability. However, the online deployment of BERT is often blocked by its large-scale parameters and high computational…

While diffusion language models (DLMs) offer a promising alternative to autoregressive models (ARs), existing open-source DLMs suffer from high inference latency. This bottleneck is mainly due to the attention's quadratic complexity with…

Recently Transformers have provided state-of-the-art performance in sparse matching, crucial to realize high-performance 3D vision applications. Yet, these Transformers lack efficiency due to the quadratic computational complexity of their…

Transformer-based QA models use input-wide self-attention -- i.e. across both the question and the input passage -- at all layers, causing them to be slow and memory-intensive. It turns out that we can get by without input-wide…

The pre-trained language models like BERT, though powerful in many natural language processing tasks, are both computation and memory expensive. To alleviate this problem, one approach is to compress them for specific tasks before…

Although Transformers have successfully transitioned from their language modelling origins to image-based applications, their quadratic computational complexity remains a challenge, particularly for dense prediction. In this paper we…

Despite the success of Deep Learning (DL), the deployment of modern DL models requiring large computational power poses a significant problem for resource-constrained systems. This necessitates building compact networks that reduce…

Diffusion Transformers (DiT) are renowned for their impressive generative performance; however, they are significantly constrained by considerable computational costs due to the quadratic complexity in self-attention and the extensive…

We introduce a simple post-training method that makes transformer attention sparse without sacrificing performance. Applying a flexible sparsity regularisation under a constrained-loss objective, we show on models up to 7B parameters that…

Transformer-based language models have found many diverse applications requiring them to process sequences of increasing length. For these applications, the causal self-attention -- which is the only component scaling quadratically w.r.t.…

We introduce EELBERT, an approach for compression of transformer-based models (e.g., BERT), with minimal impact on the accuracy of downstream tasks. This is achieved by replacing the input embedding layer of the model with dynamic, i.e.…

Sparse-Linear Attention (SLA) combines sparse and linear attention to accelerate diffusion models and has shown strong performance in video generation. However, (i) SLA relies on a heuristic split that assigns computations to the sparse or…