Finetuning Pretrained Transformers into RNNs

TL;DR

This paper introduces a method to convert pretrained transformers into efficient recurrent models by replacing their attention mechanism and finetuning, achieving better efficiency-accuracy tradeoffs without retraining from scratch.

Contribution

It proposes a swap-then-finetune procedure to transform pretrained transformers into recurrent models, enhancing efficiency while preserving accuracy.

Findings

Improved efficiency-accuracy tradeoff over standard transformers.

Lower training cost compared to training recurrent models from scratch.

Effective conversion method applicable to large-scale pretrained models.

Abstract

Transformers have outperformed recurrent neural networks (RNNs) in natural language generation. But this comes with a significant computational cost, as the attention mechanism's complexity scales quadratically with sequence length. Efficient transformer variants have received increasing interest in recent works. Among them, a linear-complexity recurrent variant has proven well suited for autoregressive generation. It approximates the softmax attention with randomized or heuristic feature maps, but can be difficult to train and may yield suboptimal accuracy. This work aims to convert a pretrained transformer into its efficient recurrent counterpart, improving efficiency while maintaining accuracy. Specifically, we propose a swap-then-finetune procedure: in an off-the-shelf pretrained transformer, we replace the softmax attention with its linear-complexity recurrent alternative and then finetune. With a learned feature map, our approach provides an improved tradeoff between efficiency and accuracy over the standard transformer and other recurrent variants. We also show that the finetuning process has lower training cost relative to training these recurrent variants from scratch. As many models for natural language tasks are increasingly dependent on large-scale pretrained transformers, this work presents a viable approach to improving inference efficiency without repeating the expensive pretraining process.