Related papers: vAttention: Dynamic Memory Management for Serving …

Multi-modal Large Language Models (MLLMs) serving systems commonly employ KV-cache compression to reduce memory footprint. However, existing compression methods introduce significant processing overhead and queuing delays, particularly in…

Large Language Models (LLMs) are increasingly deployed in scenarios demanding ultra-long context reasoning, such as agentic workflows and deep research understanding. However, long-context inference is constrained by the KV cache, a…

Role specialization in multi-LLM agent systems is often realized via multi-LoRA, where agents share a pretrained backbone and differ only through lightweight adapters. Despite sharing base model weights, each agent independently builds and…

Long-context Multimodal Large Language Models (MLLMs) that incorporate long text-image and text-video modalities, demand substantial resources as their multimodal Key-Value (KV) caches grow with increasing input lengths, challenging…

Efficient inference on GPUs using large language models remains challenging due to memory bandwidth limitations, particularly during data transfers between High Bandwidth Memory (HBM) and SRAM in attention computations. Approximate…

KV cache has traditionally been stored in GPU memory to accelerate the decoding phase of large language model (LLM) inference. However, it is increasingly necessary to move KV caches outside GPU devices, to enable cache reuse across…

The memory-for-computation paradigm of KV caching is essential for accelerating large language model (LLM) inference service, but limited GPU high-bandwidth memory (HBM) capacity motivates offloading the KV cache to cheaper external storage…

Multi-Head Attention (MHA) is a critical computational kernel in transformer-based AI models. Emerging scalable tile-based accelerator architectures integrate increasing numbers of tightly-packed processing elements (PEs) with tensor units.…

As the length of input text increases, the key-value (KV) cache in LLMs imposes prohibitive GPU memory costs and limits long-context inference on resource constrained devices. Existing approaches, such as KV quantization and pruning, reduce…

Transformer-based models have emerged as one of the most widely used architectures for natural language processing, natural language generation, and image generation. The size of the state-of-the-art models has increased steadily reaching…

KV cache restoration has emerged as a dominant bottleneck in serving long-context LLM workloads, including multi-turn conversations, retrieval-augmented generation, and agentic pipelines. Existing approaches treat restoration as a…

Multi-agent LLM systems on edge devices need to hand off latent context efficiently, but the practical choices today are expensive re-prefill or full-precision KV transfer. We study QKVShare, a framework for quantized KV-cache handoff…

Memory consumption of the Key-Value (KV) cache represents a major bottleneck for efficient large language model inference. While attention-score-based KV cache pruning shows promise, it faces critical practical limitations: attention scores…

Transformer-based models have emerged as a leading architecture for natural language processing, natural language generation, and image generation tasks. A fundamental element of the transformer architecture is self-attention, which allows…

Unified Virtual Memory (UVM) relieves the developers from the onus of maintaining complex data structures and explicit data migration by enabling on-demand data movement between CPU memory and GPU memory. However, on-demand paging soon…

Vision-Language Models (VLMs) have emerged as a critical and fast-growing extension of Large Language Models (LLMs) that enable multimodal reasoning through both text and image inputs. Although VLMs enrich the capabilities of language…

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…

Large Vision-Language Models (VLMs) have achieved remarkable success in multi-modal reasoning, but their inference time efficiency remains a significant challenge due to the memory overhead during decoding, especially when the query and…

Quarter level cell (QLC) 3D NAND flash memory is emerging as the predominant storage solution in the era of artificial intelligence. QLC 3D NAND flash stores 4 bit per cell to expand the storage density, resulting in narrower read margins.…

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…