Related papers: LayerCache: Exploiting Layer-wise Velocity Heterog…

The quadratic complexity of the attention mechanism and the substantial memory footprint of the Key-Value (KV) cache present severe computational and memory challenges for Large Language Models (LLMs) processing long contexts. Existing…

Large Language Model (LLM) inference is increasingly constrained by memory bandwidth, with frequent access to the key-value (KV) cache dominating data movement. While attention sparsity reduces some memory traffic, the relevance of past…

Recently, Flow Matching models have pushed the boundaries of high-fidelity data generation across a wide range of domains. It typically employs a single large network to learn the entire generative trajectory from noise to data. Despite…

This paper presents a general framework to build fast and accurate algorithms for video enhancement tasks such as super-resolution, deblurring, and denoising. Essential to our framework is the realization that the accuracy, rather than the…

Transformer-based large language models (LLMs) have demonstrated remarkable potential across a wide range of practical applications. However, long-context inference remains a significant challenge due to the substantial memory requirements…

Transformers have emerged as the underpinning architecture for Large Language Models (LLMs). In generative language models, the inference process involves two primary phases: prompt processing and token generation. Token generation, which…

KV cache quantization can improve Large Language Models (LLMs) inference throughput and latency in long contexts and large batch-size scenarios while preserving LLMs effectiveness. However, current methods have three unsolved issues:…

Recent work has initiated the study of dense graph processing using graph sketching methods, which drastically reduce space costs by lossily compressing information about the input graph. In this paper, we explore the strange and surprising…

Recently the generative Large Language Model (LLM) has achieved remarkable success in numerous applications. Notably its inference generates output tokens one-by-one, leading to many redundant computations. The widely-used KV-Cache…

Prefix KV caching has become a key mechanism in LLM serving: it reduces time to first token (TTFT) by avoiding redundant computation across requests that share a prefix (i.e., the system prompt). However, the accumulated KV cache is often…

Over the last few years, Deep Neural Networks (DNNs) have become ubiquitous owing to their high accuracy on real-world tasks. However, this increase in accuracy comes at the cost of computationally expensive models leading to higher…

KV caches, typically used only to speed up autoregressive decoding, encode contextual information that can be reused for downstream tasks at no extra cost. We propose treating the KV cache as a lightweight representation, eliminating the…

Network performance problems are notoriously difficult to diagnose. Prior profiling systems collect performance statistics by keeping information about each network flow, but maintaining per-flow state is not scalable on…

Aiming to minimize service delay, we propose a new random caching scheme in device-to-device (D2D)-assisted heterogeneous network. To support diversified viewing qualities of multimedia video services, each video file is encoded into a base…

Recently, large vision-language models (LVLMs) have rapidly gained popularity for their strong generation and reasoning capabilities given diverse multimodal inputs. However, these models incur significant computational and memory overhead…

Attention is the dominant source of latency during long-context LLM inference, an increasingly popular workload with reasoning models and RAG. We propose Kascade, a training-free sparse attention method that leverages known observations…

This work aims to improve the efficiency of text-to-image diffusion models. While diffusion models use computationally expensive UNet-based denoising operations in every generation step, we identify that not all operations are equally…

With the growing size of deep neural networks and datasets, the computational costs of training have significantly increased. The layer-freezing technique has recently attracted great attention as a promising method to effectively reduce…

While diffusion-based generative models have made significant strides in visual content creation, conventional approaches face computational challenges, especially for high-resolution images, as they denoise the entire image from noisy…

Diffusion models have achieved remarkable success in image and video generation tasks. However, the high computational demands of Diffusion Transformers (DiTs) pose a significant challenge to their practical deployment. While feature…