相关论文: AXELRAM: Quantize Once, Never Dequantize

The integration of Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) has significantly advanced Knowledge Graph Question Answering (KGQA). However, existing LLM-driven KGQA systems act as stateless planners, generating…

Recent advances in large language model (LLM) pretraining have led to high-quality LLMs with impressive abilities. By compressing such LLMs via quantization to 3-4 bits per parameter, they can fit into memory-limited devices such as laptops…

To help address the growing demand for ever-longer sequence lengths in transformer models, Liu et al. recently proposed Ring Attention, an exact attention algorithm capable of overcoming per-device memory bottle- necks by distributing…

Vision-language models (VLMs) such as CLIP achieve zero-shot transfer across various tasks by pre-training on numerous image-text pairs. These models often benefit from using an ensemble of context prompts to represent a class. Despite…

Normalization techniques are crucial for enhancing Transformer models' performance and stability in time series analysis tasks, yet traditional methods like batch and layer normalization often lead to issues such as token shift, attention…

The continuing advancement of memory technology has not only fueled a surge in performance, but also substantially exacerbate reliability challenges. Traditional solutions have primarily focused on improving the efficiency of protection…

In this work, we design and implement VQ-LLM, an efficient fused Vector Quantization (VQ) kernel generation framework. We first introduce a software abstraction called codebook cache to optimize codebook access efficiency and support the…

Modern large language models (LLMs) rely on system prompts to establish behavioral constraints and safety rules. Standard causal self-attention treats privileged instructions and untrusted user content with equal structural priority -- a…

Speculative decoding accelerates memory-bound LLM inference without quality degradation by using a fast drafter to propose multiple candidate tokens and the target model to verify them in parallel. However, conventional sequential…

The deployment of intelligent reinforcement learning (RL) agents on resource-constrained edge devices remains a fundamental challenge due to the substantial memory, computational, and energy requirements of modern deep learning systems.…

Quantization is the key method for reducing inference latency, power and memory footprint of generative AI models. However, accuracy often degrades sharply when activations are quantized below eight bits. Recent work suggests that…

The Segment Anything Model (SAM) achieves strong open-vocabulary segmentation, but its ViT-based image encoders dominate inference latency and memory. Existing activation compression methods, such as token merging, reduce the token length…

Processing long-context inputs with large language models presents a significant challenge due to the enormous memory requirements of the Key-Value (KV) cache during inference. Existing KV cache compression methods exhibit noticeable…

The key-value (KV) cache accelerates LLMs decoding by storing KV tensors from previously generated tokens. It reduces redundant computation at the cost of increased memory usage. To mitigate this overhead, existing approaches compress KV…

The 8 bits quantization has been widely applied to accelerate network inference in various deep learning applications. There are two kinds of quantization methods, training-based quantization and post-training quantization. Training-based…

Recent advances in Automatic Speech Recognition (ASR) have demonstrated remarkable accuracy and robustness in diverse audio applications, such as live transcription and voice command processing. However, deploying these models on…

Real-time anomaly segmentation demands both high recall and efficient low-precision inference. We study the three-way interaction of model architecture, model scale, and FP4 quantization-aware training (QAT) recipe on a recall-critical…

Processing 3D data efficiently has always been a challenge. Spatial operations on large-scale point clouds, stored as sparse data, require extra cost. Attracted by the success of transformers, researchers are using multi-head attention for…

How do we enable artificial intelligence models to improve themselves? This is central to exponentially improving generalized artificial intelligence models, which can improve their own architecture to handle new problem domains in an…

Large Language Models (LLMs) have achieved remarkable progress across reasoning, generation, and decision-making tasks, yet deploying them on mobile, embedded, and edge devices remains particularly challenging. On-device LLM inference is…