Abstract:As Pre-trained Language Models (PLMs), a popular approach for code intelligence, continue to grow in size, the computational cost of their usage has become prohibitively expensive. Prompt learning, a recent development in the field of natural language processing, emerges as a potential solution to address this challenge. In this paper, we investigate the effectiveness of prompt learning in code intelligence tasks. We unveil its reliance on manually designed prompts, which often require significant human effort and expertise. Moreover, we discover existing automatic prompt design methods are very limited to code intelligence tasks due to factors including gradient dependence, high computational demands, and limited applicability. To effectively address both issues, we propose Genetic Auto Prompt (GenAP), which utilizes an elaborate genetic algorithm to automatically design prompts. With GenAP, non-experts can effortlessly generate superior prompts compared to meticulously manual-designed ones. GenAP operates without the need for gradients or additional computational costs, rendering it gradient-free and cost-effective. Moreover, GenAP supports both understanding and generation types of code intelligence tasks, exhibiting great applicability. We conduct GenAP on three popular code intelligence PLMs with three canonical code intelligence tasks including defect prediction, code summarization, and code translation. The results suggest that GenAP can effectively automate the process of designing prompts. Specifically, GenAP outperforms all other methods across all three tasks (e.g., improving accuracy by an average of 2.13% for defect prediction). To the best of our knowledge, GenAP is the first work to automatically design prompts for code intelligence PLMs.

KiProL: A Knowledge-Injected Prompt Learning Framework for Language Generation

GRL-Prompt: Towards Knowledge Graph based Prompt Optimization via Reinforcement Learning

PromptKD: Distilling Student-Friendly Knowledge for Generative Language Models via Prompt Tuning

Genetic Auto-prompt Learning for Pre-trained Code Intelligence Language Models

Response Generation with Context-Aware Prompt Learning

RPLKG: Robust Prompt Learning with Knowledge Graph

KC-Prompt: End-To-End Knowledge-Complementary Prompting for Rehearsal-Free Continual Learning

Generating Prompts in Latent Space for Rehearsal-free Continual Learning

Infusing internalized knowledge of language models into hybrid prompts for knowledgeable dialogue generation

Editing Knowledge Representation of Language Model via Rephrased Prefix Prompts

KnowGPT: Knowledge Graph based Prompting for Large Language Models

Prompting as Probing: Using Language Models for Knowledge Base Construction

Context-Tuning: Learning Contextualized Prompts for Natural Language Generation

Knowledge Prompting in Pre-trained Language Model for Natural Language Understanding

Instance-aware Prompt Learning for Language Understanding and Generation

OpenPrompt: an Open-source Framework for Prompt-learning

KLAPrompt: Infusing Semantic Knowledge into Pre-trained Language Models by Long-answer Prompt Learning.

Adversarial Knowledge Stimulated Contrastive Prompting for Few-shot Language Learners

Dialogue for Prompting: a Policy-Gradient-Based Discrete Prompt Generation for Few-shot Learning

Genetic Prompt Search Via Exploiting Language Model Probabilities