相关论文: OSAQ: Outlier Self-Absorption for Accurate Low-bit…

Large Language Models (LLMs) have achieved state-of-the-art performance across various language tasks but pose challenges for practical deployment due to their substantial memory requirements. Furthermore, the latest generative models…

Recent works have shown a surprising result: a small fraction of Large Language Model (LLM) parameter outliers are disproportionately important to the quality of the model. LLMs contain billions of parameters, so these small fractions, such…

The emergence of accurate open large language models (LLMs) has led to a race towards performant quantization techniques which can enable their execution on end-user devices. In this paper, we revisit the problem of "extreme" LLM…

Quantization-aware training (QAT) is widely adopted to quantize language models by training full-precision weights using gradients from the quantized model. The main bottleneck is its slow convergence and early performance plateau,…

Large language models (LLMs) have achieved near-human performance across diverse reasoning tasks, yet their deployment on resource-constrained Internet-of-Things (IoT) devices remains impractical due to massive parameter footprints and…

The benefits of most large language models come with steep and often hidden economic and environmental costs due to their resource usage inefficiency during deployment. Model quantization improves energy and memory efficiency through…

Rotations have become essential to state-of-the-art quantization pipelines for large language models (LLMs) by effectively smoothing outliers in weights and activations. However, further optimizing the rotation parameters offers only…

Post-training weight-only quantization to 4 bits is widely used to reduce the memory and compute costs of large language model inference. Existing PTQ methods, such as AWQ and GPTQ, improve how weights are mapped onto a fixed 4-bit grid…

Large language models (LLMs) are omnipresent, however their practical deployment is challenging due to their ever increasing computational and memory demands. Quantization is one of the most effective ways to make them more compute and…

Outliers widely occur in big-data applications and may severely affect statistical estimation and inference. In this paper, a framework of outlier-resistant estimation is introduced to robustify an arbitrarily given loss function. It has a…

Recently years have witnessed a rapid development of large language models (LLMs). Despite the strong ability in many language-understanding tasks, the heavy computational burden largely restricts the application of LLMs especially when one…

Layer-wise quantization is a key technique for efficiently compressing large models without expensive retraining. Previous methods typically quantize the weights of each layer by "uniformly" optimizing the layer reconstruction loss across…

This work studies post-training parameter quantization in large language models (LLMs). We introduce quantization with incoherence processing (QuIP), a new method based on the insight that quantization benefits from $\textit{incoherent}$…

Post-training quantization (PTQ) has emerged as a promising technique to reduce the cost of large language models (LLMs). Specifically, PTQ can effectively mitigate memory consumption and reduce computational overhead in LLMs. To meet the…

Dynamic runtime latency and memory constraints necessitate flexible large language model (LLM) deployment, where an LLM can be inferred with various quantization precisions based on available computational resources. Recent work on such…

Post-training quantization is a key technique for reducing the memory and inference latency of large language models by quantizing weights and activations without requiring retraining. However, existing methods either (1) fail to account…

Due to the high memory and computational costs associated with large language models (LLMs), model compression techniques such as quantization, which reduces inference costs, and parameter-efficient fine-tuning (PEFT) methods like Low-Rank…

Post-training quantization of Large Language Models (LLMs) is challenging. In this work, we introduce Low-rank Quantization Error Reduction (LQER), which combines quantization and low-rank approximation to recover the model capability. LQER…

Large language models with billions of parameters are often over-provisioned: many layers contribute little unique information yet dominate the memory and energy footprint during inference. We present LieQ Layer-wise information…

Due to the dynamic nature of human language, automatic speech recognition (ASR) systems need to continuously acquire new vocabulary. Out-Of-Vocabulary (OOV) words, such as trending words and new named entities, pose problems to modern ASR…