English
Related papers

Related papers: Activation Compression in LLMs: Theoretical Analys…

200 papers

Withtherapid advancement of large language models (LLMs), the context length for inference has been continuously increasing, leading to an exponential growth in the demand for Key-Value (KV) caching. This has resulted in a significant…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-08-11 Yanyu Liu , Jingying Fu , Sixiang Liu , Yitian Zou , You Fu , Jiehan Zhou , Shouhua Zhang

Understanding the relationship between data compression and the capabilities of Large Language Models (LLMs) is crucial, especially in specialized domains like code intelligence. Prior work posited a linear relationship between compression…

Computation and Language · Computer Science 2026-03-27 Shijie Xuyang , Xianzhen Luo , Zheng Chu , Houyi Li , Siming Huang , Qiufeng Wang , Wanxiang Che , Qingfu Zhu , Shuigeng Zhou

Large Language Models (LLMs) present significant deployment challenges due to their immense size and computational requirements. Model compression techniques are essential for making these models practical for resource-constrained…

Current LLM structured pruning methods typically involve two steps: (1) compression with calibration data and (2) costly continued pretraining on billions of tokens to recover lost performance. This second step is necessary as the first…

Machine Learning · Computer Science 2024-12-31 Yaya Sy , Christophe Cerisara , Irina Illina

In this paper, we first identify activation shift, a simple but remarkable phenomenon in a neural network in which the preactivation value of a neuron has non-zero mean that depends on the angle between the weight vector of the neuron and…

Neural and Evolutionary Computing · Computer Science 2024-03-22 Takuro Kutsuna

Large multimodal models (LMMs) have advanced significantly by integrating visual encoders with extensive language models, enabling robust reasoning capabilities. However, compressing LMMs for deployment on edge devices remains a critical…

Artificial Intelligence · Computer Science 2025-07-29 Te Zhang , Yuheng Li , Junxiang Wang , Lujun Li

This study evaluates the effectiveness of zero-shot compression techniques on large language models (LLMs) under long-context. We identify the tendency for computational errors to increase under long-context when employing certain…

Computation and Language · Computer Science 2025-02-14 Chenyu Wang , Yihan Wang , Kai Li

Large language models (LLMs) excel at complex reasoning when they include intermediate steps, known as "chains of thought" (CoTs). However, these rationales are often overly verbose, even for simple problems, leading to wasted context,…

Artificial Intelligence · Computer Science 2025-07-09 Seyedarmin Azizi , Erfan Baghaei Potraghloo , Massoud Pedram

The deployment of Large Language Models is constrained by the memory and bandwidth demands of static weights and dynamic Key-Value cache. SVD-based compression provides a hardware-friendly solution to reduce these costs. However, existing…

Computation and Language · Computer Science 2026-04-03 Ruoling Qi , Yirui Liu , Xuaner Wu , Xiangyu Wang , Ming Li , Chen Chen , Jian Chen , Yin Chen , Qizhen Weng

While post-training compression techniques effectively reduce the memory footprint, latency, and power consumption of Large Language Models (LLMs), they often result in noticeable accuracy degradation and remain limited by hardware and…

Large Language Models (LLMs) demonstrate exceptional reasoning abilities, enabling strong generalization across diverse tasks such as commonsense reasoning and instruction following. However, as LLMs scale, inference costs become…

Computation and Language · Computer Science 2025-02-06 Rhea Sanjay Sukthanker , Benedikt Staffler , Frank Hutter , Aaron Klein

Leveraging large language models (LLMs) for complex natural language tasks typically requires long-form prompts to convey detailed requirements and information, which results in increased memory usage and inference costs. To mitigate these…

Computation and Language · Computer Science 2024-10-18 Zongqian Li , Yinhong Liu , Yixuan Su , Nigel Collier

Language models (LMs) excel at tasks across diverse domains, yet require substantial computational resources during inference. Compression techniques such as pruning and quantization offer a practical path towards efficient LM deployment,…

Computation and Language · Computer Science 2026-02-26 Miles Williams , George Chrysostomou , Vitor Jeronymo , Nikolaos Aletras

With the increasing adoption of Large Language Models (LLMs), more customization is needed to ensure privacy-preserving and safe generation. We address this objective from two critical aspects: unlearning of sensitive information and…

Machine Learning · Computer Science 2025-10-17 Fatmazohra Rezkellah , Ramzi Dakhmouche

Large Language Models (LLM) have achieved remarkable performance across a large number of tasks, but face critical deployment and usage barriers due to substantial computational requirements. Model compression methods, which aim to reduce…

Computation and Language · Computer Science 2025-07-15 David Ponce , Thierry Etchegoyhen , Javier Del Ser

We investigate how different compression techniques -- such as weight and activation quantization, and weight sparsity -- affect the scaling behavior of large language models (LLMs) during pretraining. Building on previous work showing that…

Machine Learning · Computer Science 2025-02-27 Elias Frantar , Utku Evci , Wonpyo Park , Neil Houlsby , Dan Alistarh

Quantization is a key method for reducing the GPU memory requirement of training large language models (LLMs). Yet, current approaches are ineffective for 4-bit activations and 8-bit gradients, which would easily cause slow convergence or…

Computation and Language · Computer Science 2026-05-12 Wenxiang Lin , Juntao Huang , Luhan Zhang , Laili Li , Xiang Bao , Mengyang Zhang , Bing Wang , Shaohuai Shi

We study COMP-AMS, a distributed optimization framework based on gradient averaging and adaptive AMSGrad algorithm. Gradient compression with error feedback is applied to reduce the communication cost in the gradient transmission process.…

Machine Learning · Statistics 2022-05-12 Xiaoyun Li , Belhal Karimi , Ping Li

Gradient aggregation has long been identified as a major bottleneck in today's large-scale distributed machine learning training systems. One promising solution to mitigate such bottlenecks is gradient compression, directly reducing…

Machine Learning · Computer Science 2024-10-30 Wenchen Han , Shay Vargaftik , Michael Mitzenmacher , Brad Karp , Ran Ben Basat

The backpropagation of error algorithm (backprop) has been instrumental in the recent success of deep learning. However, a key question remains as to whether backprop can be formulated in a manner suitable for implementation in neural…

Neural and Evolutionary Computing · Computer Science 2020-10-13 Beren Millidge , Alexander Tschantz , Anil K Seth , Christopher L Buckley