Discrete Prompt Compression with Reinforcement Learning
Authors:
Preprint
Abstract
Compressed prompts aid instruction-tuned language models (LMs) in overcoming context window limitations and reducing computational costs. Existing methods, which primarily based on training embeddings, face various challenges associated with interpretability, the fixed number of embedding tokens, reusability across different LMs, and inapplicability when interacting with black-box APIs. This study proposes prompt compression with reinforcement learning (PCRL), which is a discrete prompt compression method that addresses these issues. The proposed PCRL method utilizes a computationally efficient policy network that edits prompts directly. The training approach employed in the proposed PCRLs can be applied flexibly to various types of LMs, including both decoder-only and encoder-decoder architecture and it can be trained without gradient access to the LMs or labeled data. The proposed PCRL achieves an average reduction of 24.6\% in terms of the token count across various instruction prompts while maintaining sufficient performance. In addition, we demonstrate that the learned policy can be transferred to larger LMs, and through a comprehensive analysis, we explore the token importance within the prompts.Keywords
All Related Versions
This publication has 10 references indexed in Scilit:
- GPT understands, tooAI Open, 2024
- ChatGPT outperforms crowd workers for text-annotation tasksProceedings of the National Academy of Sciences of the United States of America, 2023
- Is ChatGPT better than Human Annotators? Potential and Limitations of ChatGPT in Explaining Implicit Hate Speech2023
- Is ChatGPT a Good NLG Evaluator? A Preliminary Study2023
- Adapting Language Models to Compress Contexts2023
- Self-Instruct: Aligning Language Models with Self-Generated Instructions2023
- On the effect of dropping layers of pre-trained transformer modelsComputer Speech & Language, 2022
- Differentiable Subset Pruning of Transformer HeadsTransactions of the Association for Computational Linguistics, 2021
- Keep It Simple: Unsupervised Simplification of Multi-Paragraph Text2021
- Ranking Sentences for Extractive Summarization with Reinforcement Learning2018