Saved in:
Bibliographic Details
Main Authors: Tsur, Dor, Adar, Sharon, Levy, Ran
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.24174
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912930990129152
author Tsur, Dor
Adar, Sharon
Levy, Ran
author_facet Tsur, Dor
Adar, Sharon
Levy, Ran
contents Small language models (SLMs) have emerged as efficient alternatives to large language models for task-specific applications. However, they are often employed in high-volume, low-latency settings, where efficiency is crucial. We propose TASC, Task-Adaptive Sequence Compression, a framework for SLM acceleration comprising two use-cases: When performing SLM fine-tuning, we propose TASC-ft, which iteratively enriches the tokenizer vocabulary with high-frequency output n-grams and then fine-tunes the model to utilize the expanded vocabulary. Next, we propose an inference-time method, termed TASC-spec. TASC-spec is a lightweight, training-free speculative decoding method that constructs an n-gram draft model from the task's output corpus, mixing task and context n-gram information.TASC-spec avoids any additional training, while bypassing draft-target vocabulary alignment constraints. We demonstrate the effectiveness of both methods across multiple low output-variability generation tasks. Our methods show consistent improvements in inference efficiency while maintaining task performance.
format Preprint
id arxiv_https___arxiv_org_abs_2602_24174
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Task-Centric Acceleration of Small-Language Models
Tsur, Dor
Adar, Sharon
Levy, Ran
Computation and Language
Artificial Intelligence
Information Theory
Small language models (SLMs) have emerged as efficient alternatives to large language models for task-specific applications. However, they are often employed in high-volume, low-latency settings, where efficiency is crucial. We propose TASC, Task-Adaptive Sequence Compression, a framework for SLM acceleration comprising two use-cases: When performing SLM fine-tuning, we propose TASC-ft, which iteratively enriches the tokenizer vocabulary with high-frequency output n-grams and then fine-tunes the model to utilize the expanded vocabulary. Next, we propose an inference-time method, termed TASC-spec. TASC-spec is a lightweight, training-free speculative decoding method that constructs an n-gram draft model from the task's output corpus, mixing task and context n-gram information.TASC-spec avoids any additional training, while bypassing draft-target vocabulary alignment constraints. We demonstrate the effectiveness of both methods across multiple low output-variability generation tasks. Our methods show consistent improvements in inference efficiency while maintaining task performance.
title Task-Centric Acceleration of Small-Language Models
topic Computation and Language
Artificial Intelligence
Information Theory
url https://arxiv.org/abs/2602.24174