Zero-Shot Continuous Prompt Transfer: Generalizing Task Semantics Across Language Models

TL;DR

This paper introduces a zero-shot method for transferring continuous prompts across different language models by encoding prompts into a relative space, enhancing cross-model generalization of task semantics.

Contribution

The work presents a novel zero-shot transfer technique for continuous prompts that encodes task semantics into a relative space, enabling effective transfer across diverse models.

Findings

Transferability of prompts improves across models

Combining multiple source prompts enhances generalization

Method confirms effectiveness through experimental results

Abstract

Prompt tuning in natural language processing (NLP) has become an increasingly popular method for adapting large language models to specific tasks. However, the transferability of these prompts, especially continuous prompts, between different models remains a challenge. In this work, we propose a zero-shot continuous prompt transfer method, where source prompts are encoded into relative space and the corresponding target prompts are searched for transferring to target models. Experimental results confirm the effectiveness of our method, showing that 'task semantics' in continuous prompts can be generalized across various language models. Moreover, we find that combining 'task semantics' from multiple source models can further enhance the generalizability of transfer.