Clip-Tuning: Towards Derivative-free Prompt Learning with a Mixture of Rewards

Understand

Derivative-free prompt learning has emerged as a lightweight alternative to prompt tuning, which only requires model inference to optimize the prompts.

• However, existing work did not take full advantage of the over-parameterized characteristics of large pre-trained language models (PLMs).
• In this paper, we propose Clip-Tuning, a simple yet effective method that adopts diverse frozen "thinned" networks of PLMs to obtain a mixture of rewards and thus advance the derivative-free prompt learning.
• The thinned networks consist of all the hidden units that survive a stationary dropout strategy, whose inference predictions reflect an ensemble of partial views over prompted training samples.