Saved in:
Bibliographic Details
Main Author: Miller, Jan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.12856
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918160779706368
author Miller, Jan
author_facet Miller, Jan
contents The Efficient Adaptive Transformer (EAT) framework unifies three adaptive efficiency techniques - progressive token pruning, sparse attention, and dynamic early exiting - into a single, reproducible architecture for input-adaptive inference. EAT provides an open-source benchmarking pipeline that automates data processing, timing, and ablation across GLUE tasks (SST-2, QQP, MNLI). Although this empirical study finds that combining these mechanisms can increase latency in shallow six-layer models, it demonstrates that EAT achieves slightly higher accuracy than the optimized DistilBERT baseline on SST-2, illustrating the potential of dynamic computation for latency-sensitive NLP. The main contribution is the open, end-to-end reproducible framework - complete with scripts, CSV logging, and analysis utilities - intended to serve as a community tool for further research on adaptive transformers.
format Preprint
id arxiv_https___arxiv_org_abs_2510_12856
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Efficient Adaptive Transformer: An Empirical Study and Reproducible Framework
Miller, Jan
Computation and Language
Artificial Intelligence
The Efficient Adaptive Transformer (EAT) framework unifies three adaptive efficiency techniques - progressive token pruning, sparse attention, and dynamic early exiting - into a single, reproducible architecture for input-adaptive inference. EAT provides an open-source benchmarking pipeline that automates data processing, timing, and ablation across GLUE tasks (SST-2, QQP, MNLI). Although this empirical study finds that combining these mechanisms can increase latency in shallow six-layer models, it demonstrates that EAT achieves slightly higher accuracy than the optimized DistilBERT baseline on SST-2, illustrating the potential of dynamic computation for latency-sensitive NLP. The main contribution is the open, end-to-end reproducible framework - complete with scripts, CSV logging, and analysis utilities - intended to serve as a community tool for further research on adaptive transformers.
title Efficient Adaptive Transformer: An Empirical Study and Reproducible Framework
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2510.12856