Efficient Large Language Models with Zero-Shot Adjustable Acceleration
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916936924790784 |
|---|---|
| author | Kachuee, Sajjad Sharifkhani, Mohammad |
| author_facet | Kachuee, Sajjad Sharifkhani, Mohammad |
| contents | Using Large Language Models (LLMs) in real-world applications presents significant challenges, particularly in balancing computational efficiency with model performance. Optimizing acceleration after fine-tuning and during inference is critical for building efficient architectures. This paper introduces Zero-Shot Adjustable Acceleration, a novel training and inference method that dynamically adjusts hardware utilization during inference without requiring additional fine-tuning. The proposed approach is applied to recent LLMs and evaluated across multiple classification and text generation tasks. Experimental results demonstrate that the method supports a wide range of zero-shot acceleration and achieves up to 11x speedup compared to the baseline. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_01190 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Efficient Large Language Models with Zero-Shot Adjustable Acceleration Kachuee, Sajjad Sharifkhani, Mohammad Computation and Language Using Large Language Models (LLMs) in real-world applications presents significant challenges, particularly in balancing computational efficiency with model performance. Optimizing acceleration after fine-tuning and during inference is critical for building efficient architectures. This paper introduces Zero-Shot Adjustable Acceleration, a novel training and inference method that dynamically adjusts hardware utilization during inference without requiring additional fine-tuning. The proposed approach is applied to recent LLMs and evaluated across multiple classification and text generation tasks. Experimental results demonstrate that the method supports a wide range of zero-shot acceleration and achieves up to 11x speedup compared to the baseline. |
| title | Efficient Large Language Models with Zero-Shot Adjustable Acceleration |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.01190 |