Related papers: FlowKV: A Disaggregated Inference Framework with L…

We present our framework DKVF that enables one to quickly prototype and evaluate new protocols for key-value stores and compare them with existing protocols based on selected benchmarks. Due to limitations of CAP theorem, new protocols must…

Large language model (LLM) based agentic workflows have become a popular paradigm for coordinating multiple specialized agents to solve complex tasks. To improve serving efficiency, existing LLM systems employ prefix caching to reuse…

In Large Language Model (LLM) inference, Key-Value (KV) caches (KV-caches) are essential for reducing time complexity. However, they result in a linear increase in GPU memory as the context length grows. While recent work explores KV-cache…

RWKV is a modern RNN architecture that approaches the performance of Transformers, with the advantage of processing long contexts at a linear memory cost. However, its sequential computation pattern struggles to efficiently leverage GPU…

While Key-Value (KV) cache compression is essential for efficient LLM inference, current evaluations disproportionately focus on sparse retrieval tasks, potentially masking the degradation of High-Density Reasoning where Chain-of-Thought…

Large Language Models (LLMs) have achieved impressive accomplishments in recent years. However, the increasing memory consumption of KV cache has possessed a significant challenge to the inference system. Eviction methods have revealed the…

We observe two major trends in LLM-based generative AI: (1) inference is becoming the dominant factor in terms of cost and power consumption, surpassing training, and (2) retrieval augmented generation (RAG) is becoming prevalent. When…

Large language models (LLMs) are increasingly deployed in AI infrastructure, driving the need for high throughput, resource efficient serving systems. Disaggregated LLM serving, which separates prompt prefill from auto-regressive decode,…

This work proposes an efficient method to enhance the quality of corrupted speech signals by leveraging both acoustic and visual cues. While existing diffusion-based approaches have demonstrated remarkable quality, their applicability is…

Distributed LLM serving is costly and often underutilizes hardware accelerators due to three key challenges: bubbles in pipeline-parallel deployments caused by the bimodal latency of prompt and token processing, GPU memory overprovisioning,…

Disaggregated LLM serving improves resource efficiency by separating the compute-intensive prefill phase from the latency-critical decode phase. However, this architecture introduces a fundamental bottleneck: key/value (KV) tensors…

The key-value (KV) cache is a major bottleneck in long-context inference, where memory and computation grow with sequence length. Existing KV eviction methods reduce this cost but typically degrade performance relative to full-cache…

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…

We propose ReKV, a novel training-free approach that enables efficient streaming video question-answering (StreamingVQA), by seamlessly integrating with existing Video Large Language Models (Video-LLMs). Traditional VideoQA systems struggle…

KV cache eviction has emerged as an effective solution to alleviate resource constraints faced by LLMs in long-context scenarios. However, existing token-level eviction methods often overlook two critical aspects: (1) their irreversible…

Multimodal large language models suffer from substantial inference overhead since multimodal KV Cache grows proportionally with the visual input length. Existing multimodal KV Cache compression methods mostly rely on attention score to…

Context lengths of Large Language Models (LLMs) have exploded in recent years, with 128k-token context becoming a standard and million-token context becoming a reality. Efficiently supporting long-context inference remains challenging as…

Attentio-FFN disaggregation (AFD) is an emerging architecture for LLM decoding that separates state-heavy, KV-cache-dominated Attention computation from stateless, compute-intensive FFN computation, connected by per-step communication.…

In this study, we introduce adaptive KV cache compression, a plug-and-play method that reduces the memory footprint of generative inference for Large Language Models (LLMs). Different from the conventional KV cache that retains key and…

We present PolyKV, a system in which multiple concurrent inference agents share a single, asymmetrically compressed KV cache pool. Rather than allocating a separate KV cache per agent -- the standard paradigm -- PolyKV writes a compressed…