相关论文: Breaking the KV Cache Bottleneck: Fan Duality Mode…

Federated learning (FL) has shown great potential in medical image computing since it provides a decentralized learning paradigm that allows multiple clients to train a model collaboratively without privacy leakage. However, current studies…

Precise estimation of model inference latency is crucial for time-critical mobile edge applications, enabling devices to calculate latency margins against deadlines and trade them for enhanced model performance or resource savings. However,…

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…

This paper uses the reconstruction-computation-quantization (RCQ) paradigm to decode low-density parity-check (LDPC) codes. RCQ facilitates dynamic non-uniform quantization to achieve good frame error rate (FER) performance with very low…

Deep learning-based Visual SLAM (vSLAM) systems exhibit exceptional geometric reasoning capabilities, yet their prohibitive computational overhead severely restricts deployment on resource-constrained autonomous platforms. This paper…

Mechanistic interpretability aims to reverse-engineer the internal computations of Large Language Models (LLMs), yet separating sparse semantic signals from high-dimensional polysemantic noise remains a significant challenge. This paper…

LLM decoding is bottlenecked for large batches and long contexts by loading the key-value (KV) cache from high-bandwidth memory, which inflates per-token latency, while the sequential nature of decoding limits parallelism. We analyze the…

To address the increasing computational demands of artificial intelligence (AI) and big data, compute-in-memory (CIM) integrates memory and processing units into the same physical location, reducing the time and energy overhead of the…

Federated Learning (FL) enables multiple devices to collaboratively train a shared model while preserving data privacy. Ever-increasing model complexity coupled with limited memory resources on the participating devices severely bottlenecks…

Multi-agent LLM systems on edge devices face a memory management problem: device RAM is too small to hold every agent's KV cache simultaneously. On Apple M4 Pro with 10.2 GB of cache budget, only 3 agents fit at 8K context in FP16. A…

In this work, an efficient precoding design scheme is proposed for downlink cell-free distributed massive multiple-input multiple-output (DM-MIMO) filter bank multi-carrier (FBMC) systems with asynchronous reception and highly frequency…

As Large Language Models (LLMs) scale in size and context length, the memory requirements of the key value (KV) cache have emerged as a major bottleneck during autoregressive decoding. The KV cache grows with sequence length and embedding…

The Transformer architecture, underpinned by the self-attention mechanism, has become the de facto standard for sequence modeling tasks. However, its core computational primitive scales quadratically with sequence length (O(N^2)), creating…

Cyber-security for 5G networks is drawing notable attention due to an increase in complex jamming attacks that could target the critical 5G Radio Frequency (RF) domain. These attacks pose a significant risk to heterogeneous network (HetNet)…

The popular VQ-VAE models reconstruct images through learning a discrete codebook but suffer from a significant issue in the rapid quality degradation of image reconstruction as the compression rate rises. One major reason is that a higher…

Federated Learning (FL) is an emerging framework for distributed processing of large data volumes by edge devices subject to limited communication bandwidths, heterogeneity in data distributions and computational resources, as well as…

Large language models (LLMs) are typically served from clusters of GPUs/NPUs that consist of large number of devices. Unfortunately, communication between these devices incurs significant overhead, increasing the inference latency and cost…

As modern LLMs support thousands to millions of tokens, KV caches grow to hundreds of gigabytes, stressing memory capacity and bandwidth. Existing solutions, such as KV cache pruning and offloading, alleviate these but underutilize hardware…

Multimodal Large Language Models (MLLMs) have advanced unified reasoning over text, images, and videos, but their inference is hindered by the rapid growth of key-value (KV) caches. Each visual input expands into thousands of tokens,…

We introduce the Holographic Knowledge Manifold (HKM), a four-phase pipeline that achieves zero catastrophic forgetting in AI knowledge representation while maintaining minimal memory growth and high efficiency. Leveraging fractal…