LS-PRISM:一种通过低级近似和散散的层选择性修剪方法,用于高效的大型语言模型压缩
Renshuai Tao1, Hairong Chen1, Yuzhe Guo1
1Beijing Jiaotong University, Beijing, 100044, China.
概括
我们开发了LS-PRISM,这是一种通过选择性修剪层来压缩大型语言模型 (LLM) 的新方法. 这种技术可以显著减少模型大小,同时在NLP任务上保持高性能.
科学领域:
- 人工智能的人工智能
- 自然语言处理自然语言处理.
- 机器学习 机器学习
背景情况:
- 大型语言模型 (LLM) 在自然语言处理 (NLP) 中实现了最先进的性能.
- 对于资源有限的环境来说,LLM的实质性参数数量带来了部署挑战.
- 现有的压缩方法经常在所有层中应用均的压缩,可能会不均地影响性能.
研究的目的:
- 引入LS-PRISM,一种新的通过低级近似和稀疏化方法进行层选择性修剪的新方法.
- 为了有效地压缩LLM,同时保持关键NLP基准的性能.
- 为在资源有限的环境中部署LLM提供可扩展的解决方案.
主要方法:
- LS-PRISM采用基于精度和损失影响的层选择性低级近似.
- 动态排名选择可自适应地确定近似排名,以保持最佳性能.
- 非结构化的修剪和可选的LoRA微调进一步增强了模型散射和性能恢复.
主要成果:
- 在参数计数和存储方面实现了显著的减少.
- 在NLP基准 (BoolQ,RTE,ARC-Challenge) 中观察到最小的精度下降.
- 在具有可比性能的2.5B参数LLM上显示了高达12%的参数减少.
结论:
- LS-PRISM提供了一种有效和可扩展的方法来压缩LLMs.
- 该方法成功地平衡了模型压缩与性能保存.
- LS-PRISM适用于在资源有限的环境中部署LLM.
相关概念视频
Improving Translational Accuracy
11.9K
Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...
11.9K
Language Development
454
Children master language quickly and with relative ease, supported by both biological predisposition and reinforcement. B. F. Skinner (1957) proposed that language is learned through reinforcement, while Noam Chomsky (1965) argued that language acquisition mechanisms are biologically determined.
The critical period for language acquisition suggests that the ability to acquire language is at its peak early in life. As people age, this proficiency decreases. Language development begins very...
The critical period for language acquisition suggests that the ability to acquire language is at its peak early in life. As people age, this proficiency decreases. Language development begins very...
454
Reducing Line Loss
194
In a three-phase circuit, line loss is an indicator of energy dissipated as heat due to the resistance of transmission lines. To address this, incorporating transformers into the system—a step-up transformer at the source and a step-down transformer at the load—is a strategic solution. Two three-phase transformers are introduced to improve this.
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...
With a step-up transformer at the source, the voltage is increased, thereby reducing the current in the transmission lines since power loss...
194
Language and Cognition
444
Language serves as a bridge between ideas and communication, influencing how individuals perceive and interact with the world. Psychologists have long debated whether language shapes thought or vice versa. This discussion gained grip with Edward Sapir and Benjamin Lee Whorf in the 1940s, who proposed that language determines thought, a concept known as linguistic determinism. They suggested that the vocabulary and structure of a language influence how its speakers think and perceive reality.
444
One-Compartment Open Model: Wagner-Nelson and Loo Riegelman Method for ka Estimation
717
This lesson introduces two critical methods in pharmacokinetics, the Wagner-Nelson and Loo-Riegelman methods, used for estimating the absorption rate constant (ka) for drugs administered via non-intravenous routes. The Wagner-Nelson method relates ka to the plasma concentration derived from the slope of a semilog percent unabsorbed time plot. However, it is limited to drugs with one-compartment kinetics and can be impacted by factors like gastrointestinal motility or enzymatic degradation.
On...
On...
717
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
101
Mechanistic models play a crucial role in algorithms for numerical problem-solving, particularly in nonlinear mixed effects modeling (NMEM). These models aim to minimize specific objective functions by evaluating various parameter estimates, leading to the development of systematic algorithms. In some cases, linearization techniques approximate the model using linear equations.
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
101


