Advances in Software Development with Large Language Models

The field of software development is witnessing significant advancements with the integration of Large Language Models (LLMs). Researchers are exploring the potential of LLMs in various aspects of software development, including issue resolution, requirements elicitation, and sentiment analysis. The use of LLMs is shown to improve the efficiency and effectiveness of these processes, enabling developers to focus on higher-level tasks. Notably, LLMs are being leveraged to automate tasks such as generating privacy requirements, creating user scenarios, and conducting requirements elicitation interviews. Overall, the field is moving towards increased adoption of LLMs to enhance software development workflows. Noteworthy papers in this area include:

  • a study on the effectiveness of ChatGPT for software issue resolution, which found that ChatGPT is most effective for code generation and tools/libraries/APIs recommendations.
  • a paper on generating privacy stories from software documentation, which developed a novel approach based on chain-of-thought prompting and in-context-learning to extract privacy behaviors from software documents.

Sources

What Makes ChatGPT Effective for Software Issue Resolution? An Empirical Study of Developer-ChatGPT Conversations in GitHub

Generating Privacy Stories From Software Documentation

If You Had to Pitch Your Ideal Software -- Evaluating Large Language Models to Support User Scenario Writing for User Experience Experts and Laypersons

Towards Trustworthy Sentiment Analysis in Software Engineering: Dataset Characteristics and Tool Selection

LLMREI: Automating Requirements Elicitation Interviews with LLMs

Legal Requirements Translation from Law

Requirements Elicitation Follow-Up Question Generation

Built with on top of