Results of the summarizing process for the arXiv paper: 2405.03256v1

, , , , In the field of software development, Requirements Engineering (RE) plays a crucial role in translating stakeholders' needs into detailed requirements specifications. Recent advancements in deep learning techniques have shown promise in enhancing various RE tasks. However, effective collaboration across different tasks and roles involved in the process is necessary to achieve high-quality requirements specifications. To address this challenge, a novel framework called MARE (Multi-Agents Collaboration Framework for Requirements Engineering) has been proposed. MARE leverages the power of large language models (LLMs) to facilitate collaboration throughout the entire RE process. The framework divides the RE process into four key tasks: elicitation, modeling, verification, and specification. Each task is assigned to one or two specific agents, with each agent capable of performing multiple actions. In total, MARE consists of five agents and nine distinct actions. One of the key features of MARE is the provision of a dedicated workspace where agents can upload their generated intermediate requirements artifacts and access necessary information for seamless collaboration. To evaluate the effectiveness of MARE, experiments were conducted on a variety of public cases as well as new cases developed specifically for this study. A comparison with three baseline approaches using commonly used metrics revealed that MARE outperformed existing methods by 15.4% in generating correct requirements models. Furthermore, human evaluations were conducted to assess the quality of the requirements specifications generated by MARE across three aspects. The insights gained from these evaluations provide valuable feedback on the performance and capabilities of the framework in producing high-quality requirements specifications. Overall, MARE represents a significant advancement in enhancing collaboration among agents involved in Requirements Engineering processes. It demonstrates superior performance compared to state-of-the-art approaches and offers valuable insights into improving the quality of generated requirements models and specifications.

• Requirements Engineering (RE) is crucial for translating stakeholders' needs into detailed requirements specifications
• Recent advancements in deep learning techniques show promise in enhancing various RE tasks
• Effective collaboration across different tasks and roles is necessary for high-quality requirements specifications
• MARE (Multi-Agents Collaboration Framework for Requirements Engineering) leverages large language models to facilitate collaboration in the RE process
• MARE divides the RE process into four key tasks: elicitation, modeling, verification, and specification, each assigned to specific agents capable of multiple actions
• MARE consists of five agents and nine distinct actions, providing a dedicated workspace for seamless collaboration
• Experiments show that MARE outperformed existing methods by 15.4% in generating correct requirements models
• Human evaluations confirmed the quality of requirements specifications generated by MARE across three aspects
• MARE represents a significant advancement in enhancing collaboration among agents involved in Requirements Engineering processes

Introduction

Requirements Engineering (RE) is a critical process in software development that involves translating stakeholders' needs into detailed requirements specifications. It plays a crucial role in ensuring the success of software projects by bridging the gap between what users want and what developers deliver. However, RE is a complex and challenging task that requires effective collaboration among different roles and tasks involved in the process.

In recent years, deep learning techniques have shown great potential in enhancing various RE tasks such as requirements elicitation, modeling, verification, and specification. These techniques use large language models (LLMs) to analyze natural language text and generate high-quality requirements artifacts. However, there is still room for improvement when it comes to collaboration across different tasks and roles within the RE process.

To address this challenge, researchers have proposed a novel framework called MARE (Multi-Agents Collaboration Framework for Requirements Engineering). This framework leverages the power of LLMs to facilitate collaboration throughout the entire RE process. In this article, we will provide an overview of MARE and discuss its key features and capabilities.

The MARE Framework

MARE divides the RE process into four key tasks: elicitation, modeling, verification, and specification. Each task is assigned to one or two specific agents with unique capabilities. The framework consists of five agents: Elicitation Agent (EA), Modeling Agent (MA), Verification Agent (VA), Specification Agent 1 (SA1), and Specification Agent 2 (SA2). Each agent can perform multiple actions related to their assigned task(s).

The EA's main function is to gather information from stakeholders through interviews or surveys using natural language processing techniques. The MA uses this information to create initial requirements models using LLMs such as BERT or GPT-3. The VA then verifies these models against predefined constraints or rules provided by domain experts.

Once verified, SA1 generates intermediate requirements specifications, which are then reviewed and refined by SA2. The final output of MARE is a set of high-quality requirements models and specifications that can be used by developers to build the software.

Collaboration Workspace

One of the key features of MARE is the provision of a dedicated workspace where agents can upload their generated artifacts and access necessary information for seamless collaboration. This workspace serves as a central hub for all agents involved in the RE process, allowing them to share their work and communicate effectively.

The workspace also provides real-time updates on the progress of each task, ensuring that all agents are aware of any changes or updates made by others. This promotes transparency and reduces the chances of miscommunication or conflicting requirements.

Evaluation Results

To evaluate the effectiveness of MARE, experiments were conducted on both public cases and new cases developed specifically for this study. Three baseline approaches commonly used in RE were compared with MARE using metrics such as precision, recall, F1-score, and accuracy.

The results showed that MARE outperformed existing methods by 15.4% in generating correct requirements models. It also achieved higher scores in terms of precision, recall, F1-score, and accuracy compared to the baselines.

Furthermore, human evaluations were conducted to assess the quality of requirements specifications generated by MARE across three aspects: completeness (whether all necessary information was included), consistency (whether there were any conflicts or contradictions), and clarity (how easy it was to understand). The results showed that MARE scored significantly higher than existing methods in all three aspects.

Conclusion

MARE represents a significant advancement in enhancing collaboration among agents involved in Requirements Engineering processes. It leverages LLMs to facilitate communication and cooperation throughout the entire RE process while providing a dedicated workspace for seamless collaboration.

The evaluation results demonstrate its superior performance compared to state-of-the-art approaches and offer valuable insights into improving the quality of generated requirements models and specifications. MARE has the potential to greatly improve the efficiency and effectiveness of RE, ultimately leading to better software products that meet stakeholders' needs.