The Power of Scale for Parameter-Efficient Prompt Tuning
This addresses the high cost of sharing and serving large models by enabling reuse of a single frozen model for multiple tasks, though it builds on prior work like prefix tuning.
The paper tackles the problem of adapting large language models to downstream tasks efficiently by introducing prompt tuning, a method that learns soft prompts for frozen models, and shows it outperforms GPT-3's few-shot learning and matches full model tuning performance at scale, with ablations on T5 models exceeding billions of parameters demonstrating this gap closure.
In this work, we explore "prompt tuning", a simple yet effective mechanism for learning "soft prompts" to condition frozen language models to perform specific downstream tasks. Unlike the discrete text prompts used by GPT-3, soft prompts are learned through backpropagation and can be tuned to incorporate signal from any number of labeled examples. Our end-to-end learned approach outperforms GPT-3's "few-shot" learning by a large margin. More remarkably, through ablations on model size using T5, we show that prompt tuning becomes more competitive with scale: as models exceed billions of parameters, our method "closes the gap" and matches the strong performance of model tuning (where all model weights are tuned). This finding is especially relevant in that large models are costly to share and serve, and the ability to reuse one frozen model for multiple downstream tasks can ease this burden. Our method can be seen as a simplification of the recently proposed "prefix tuning" of Li and Liang (2021), and we provide a comparison to this and other similar approaches. Finally, we show that conditioning a frozen model with soft prompts confers benefits in robustness to domain transfer, as compared to full model tuning.